УДК 811.111ББК 81.432.1

Р е ц е н з е н т ы:

Власюк Н.И., кандидат филологических наук, доцент;

Гущина Л.Н., доцент кафедры иностранных языков,кандидат филологических наук (ГГМУ).

Рекомендовано Советом филологического факультетаГрГУ им. Я. Купалы.

Малышева,О.Л.Теоретическая грамматика английского языка : в 2 ч. Ч 1:

пособие / О.Л. Малышева, Е.О. Мочалова. – Гродно :ГрГУ, 2009. – 145 с.

ISBN 978-985-515-152-5

Излагается системное описание грамматического строя английского языка,приведены практические задания и тесты, задания к семинарским занятиям, вы-держки из работ лингвистов по спорным вопросом теоритической грамматики. Ад-ресуется студентам специальностей: «Современные иностранные языки (препода-вание)», «Лингвистическое обеспечение межкультурных коммуникаций (междуна-родный туризм)», «Английский язык. Немецкий язык», «Английский язык.Французский язык».

УДК 811.111ББК 81.432.1

М20

М20

© Малышева О.Л., Мочалова Е.О., 2009© Учреждение образования «Гродненский государственный университет имени Янки Купалы», 2009ISBN 978-985-515-152-5

3

CONTENTS

Предисловие ......................................................................................5

1. History of Grammar ......................................................................61.1. Introduction ...................................................................................61.2. Traditional Grammar .....................................................................9

1.2.1. Traditional Grammar in Ancient Greece .............................91.2.2. Traditional Grammar in Ancient Rome .............................131.2.3. Traditional Grammar in Ancient India ..............................17

Check yourself test 1 ..........................................................................201.3. Prescriptive Grammar .................................................................221.4. The Historical Comparative Method in Linguistics ...................241.5. Non-Structural Descriptive Grammar .........................................29Check yourself test 2 ..........................................................................301.6. The Fundamental Ideas and Main Schoolsof Modern Linguistics ........................................................................32Check yourself test 3 ..........................................................................391.7. Structural Descriptive Grammar .................................................42Check yourself test 4 ..........................................................................501.8. Transformational-Generative Grammar ......................................511.9. Contemporary Descriptive Linguistics .......................................551.10. The Explanatory Power of Non-Structural Descriptive,Structural Descriptive and Transformational-Generative Grammar ....58

1.10.1. Non-Structural Descriptive Grammar .............................581.10.2. Structural Descriptive Grammar ......................................601.10.3. Transformational-Generative Grammar ..........................61

2. Grammar: Theoretical Issues .....................................................632.1. Practical and Theoretical Grammar ............................................632.2. Paradigmatic and Syntagmatic Relations ....................................632.3. Language and Speech ..................................................................692.4. Language Units and Language Levels ........................................70

4

3. Morphemic Structure of the Word .............................................783.1. Morphology and Morpheme .......................................................783.2. Special nature of the Morpheme .................................................793.3. Word as a Nominative Unit .........................................................823.4. Morphemic Structure of the Word ..............................................833.5. Morphological Processes ............................................................84

3.5.1. Processes affecting Grammatical Form .............................843.5.2. Processes affecting Grammatical Content .........................863.6.1. The Analytic and Synthetic Forms ....................................923.6.1. Analytic Forms ...................................................................93

4. Categorization in Morphology ....................................................964.1. Notion of Opposition. Oppositions in Morphology ....................964.2. Oppositional Reduction ...............................................................974.3. Categorization: Categories, Categorial Forms ............................984.4. Grammatical Categories ............................................................1014.5. The Notion of Grammatical Category .......................................103

Selected Reader ..............................................................................109

The List of Sources .........................................................................142

Recommended Literature .............................................................144

5

ПРЕДИСЛОВИЕ

Предлагаемое пособие предназначено для студентов филологи-ческого факультета университета в качестве руководства для лекцион-ных и практических занятий по курсу теоретической грамматики анг-лийского языка. Оно ориентировано на системное описание граммати-ческого строя английского языка. Данное пособие является первойчастью учебно-методического комплекса по теоретической граммати-ке английского языка.

Курс теоретической грамматики английского языка входит в чис-ло базовых курсов обучения студентов, для которых английский языкявляется профилирующей дисциплиной специальности.

Отвечая требованиям программы, курс ставит своей целью, вме-сте с комплексным описанием грамматического строя английского язы-ка, дать обобщающее введение в проблематику современных грамма-тических исследований и, соответственно, в методику научно-грамма-тического анализа языкового материала.

Непосредственной учебной задачей курса является:1. Теоретическое освещение основы грамматического строя анг-

лийского языка в соответствии с современным состоянием науки о язы-ке в его двух взаимосвязанных и взаимодополнительных функциях –когнитивной и коммуникативной.

2. Ввести студентов в наиболее важные проблемы современныхнаучных исследований грамматического строя английского языка.

3. Развивать у студентов умение применять теоретические знанияпо грамматике языка к практическому преподаванию английского язы-ка на разных ступенях обучения.

4. Развить у студентов научное мышление, соответствующее методо-логии предмета теоретической грамматики, научить их библиографическо-му поиску в изучаемой области, привить им умение самостоятельно пере-рабатывать фундаментальную и текущую научную информацию по пред-мету, самостоятельно делать обобщения и выводы из данных, приводимыхв специальной литературе, а также из собственных наблюдений над факти-ческим языковым материалом в его разных речевых формах, осмысленносопоставлять грамматические явления английского и родного языков.

Важно учитывать, что введение студентов в проблемно-дискус-сионную часть курса должно быть нацелено на демонстрацию общеговосхождения науки в познании предмета восхождения, которое осу-ществляется именно в результате раскрытия проблем, возникающих врезультате различного истолкования реальных, диалектически слож-ных и противоречивых языковых явлений.

Для контроля самостоятельной работы студентов предлагаются тес-ты, которые в зависимости от уровня подготовки студентов могут выпол-няться либо на практических занятиях, либо во внеаудиторное время.

6

1. HISTORY OF GRAMMAR

1.1. INTRODUCTION

The term grammar is derived from the Greek word grammatikз,where gram means ‘something written’. The part tikз derives from tech-nз and means ‘art’. Hence grammatikз is the art of writing. Since itsappearance in ancient Greece the term has undergone considerable mod-ifications. In ancient Greece and ancient Rome the terms grammatikзand grammatica respectively denoted the whole apparatus of literarystudy. In the middle ages, grammar was the study of Latin.

In England, this conception of grammar continued until the end ofthe 16th century. Latin grammar was the only grammar learned in schools.Until then there were no grammars of English. The first grammar of En-glish, Brief Grammar of English, written by William Bullokar, was pub-lished in 1585. The most influential grammar of English (published in1762) was R. Lowth’s Short Introduction to English Grammar. It startedthe age of prescriptive grammar.

Historically, English grammars, according to their general aims andobjectives, can be divided into: a) traditional (prescriptive and non-struc-tural descriptive): b) structural descriptive and c) transformational-genera-tive. To a prescriptive grammarian, grammar is rules of correct usage; itsaim is to prescribe what is judged to be correct rather than to describeactual usage. A new, modern understanding of grammar appeared only bythe end of the 19th century, when the period of scientific (descriptive) gram-mar began. To descriptivists, grammar is a systematic description of thestructure of a language. With the appearance of structural descriptive lin-guistics, grammar came to mean the system of word structures and wordarrangements of a given language at a given time. To transformational-generative grammarians, who are an off-shoot of structural descriptivelinguistics, grammar is a mechanism for producing sentences.

According to R.P. Milrud "the prescriptive approach usually emphasisesgrammar prototypes, i.e. the most typical cases of language use that can beexplained by consistent rules. Yet the actual language use and the tendencies inthe development of the language do not fit completely in the set of prototypes.E.g. sentences like "There's so many good shops in the city centre", or "There'smany more lights on the other street" appear to be relatively stable in the spo-ken norm of the modern language though not recognized by the grammar books.

Another example of the kind is the "indirect questions" such as "Doyou want to know where's your mother?". Sequence of tenses is getting

7

more meaningful than formal in the English sentences and this tendencyis already registered in the tendencies for teaching grammar. E.g. "Hesaid he would celebrate the Millennium in San Francisco" is a grammat-ical possibility conveying the idea that the travel to the city is still ahead.

The prescriptive approach makes language users form comparativeand superlative degrees of one/two syllable adjectives with the suffixes"-er" and "-est", e.g. "wise – wiser – the wisest". Language reality seemsto violate this rule allowing for cases such as "clever – more clever – themost clever", "fresh – more fresh – the most fresh" etc.

The use of grammar tenses is another controversial issue. E.g. PresentProgressive is said to express an action at the moment of speech. Thisrule does not account for cases such as "The team is playing for the Arse-nal this season", or "I am baby-sitting these days". Neither "the momentof speech", nor "action in progress" can be applied in the above cases.

According to normative grammar, Past Perfect Tense is used to ex-press priority to another action in the past. In other words it is used when itis necessary to emphasise that the action expressed with the Past Perfecttook place prior to the past action that took place later. Past Perfect Tenseis contrasted to Past Simple Tense that is used to denote the actions thattook place in the past with a worded or implied time marker, such as "in1970". Language reality seems to overrule this assumption. Let's considerthe following examples, "1)I became Managing Director five years ago.2) I had been Personnel Manager for three years and 3) I joined/had joinedthe firm in 1970". Sentence 1 does not contradict the rules of the normativegrammar. Sentence 2 would be normally said in Past Simple as there is animplied time marker "then". Sentence 3 would definitely be used in PastSimple as there is a verbalised time marker (in 1970). Yet, according to thesuggested interpretation, Past Perfect can be used with time markers suchas "five years ago" and can answer the question "When?". In contrast tothe Past Perfect, Present Perfect is said to refer to the moment of speech.It cannot answer the question "When?" At the same time, adverbs suchas "just now" and "recently" can answer the question "When?". Thismakes sentences such as "He told me just now that he only recently be-came Managing Director" quite correct.

A controversial issue for teaching English grammar is the use of thePresent or Future Simple Tenses in the "if" and "when" clauses of "con-dition" and "time". The regular practice is to use the Present Simple inorder to communicate futurity and all the instances where "will" is usedafter "if" and "when" are viewed by many teachers with caution as dis-putable language use.

It is absolutely true that in many cases English language speakersprefer Present Simple in order to describe future after "if" and "when".

8

This is a language practice going back to the times of Old English whenthe category of the future was expressed often lexically with the verbbeing used in the Present Tense, e.g. "Weird saves oft the man undoomedif he undaunted be!" – is pronounced by Beowulf as a prediction beforehe sets out to fight seven sea-monsters. Present Simple was used in theclassical "Beowulf and Grendel" to express the future in simple sentenc-es such as "Back at the sea-shore I resume the watch against sea-raiders."

Modal meaning of the future in the "if" and "when" clauses is wide-ly used by Shakespeare, i.e. at the time when the general features ofModern English had taken shape, "When forty winters shall besiege thybrow...", "Then what could death do if though shouldst depart". At thesame time, the loss of the Future Simple after "if" was also not excludedand in fact typical, "If thow survive my well contented day...", "If for mylove thou my love receivest...", "But yet be blamed if thou thyself de-ceivest..." (the data is from the concordance of Shakespeare's sonnets).

One can assume that the use of "will" and "shall" in the condition/time clauses has not been uncommon for the English language during itshistorical existence and this tendency develops further on with new modalmeanings other than "volition" and "declaration" being added to the gamut,e.g. "If the train will arrive so late, let's book a taxi" (strong probability).Regarding the use of Present Simple to express futurity in Modern En-glish, one can also conclude that this is not a peculiar and uncomfortableexception to the general rule, but rather the survival of the Old Englishtendency to express futurity with the verbs in the Present Tense.

Normative grammar attempted to dictate word formation as well.E.g. Shakespeare's words such as "laughable" ought to have been changedto "laughatable". The word "reliable" was to be pronounced and spelt as"relionable" etc. Actually, what is found in Shakespeare's language is byno means a poetic twist of the language or the violation of the norm topoetic ends. Shakespeare's language is a fair mirror of living Englishwith many tendencies surviving to date and described in the grammarbooks for advanced students.

A vivid example of prescriptive linguistics is the use of the definiteand indefinite articles. According to the most general rules, the indefi-nite article is used when an object is mentioned as the first occurrence inthe text and we know little if anything about it. The definite article isused in cases, when the object is known to the speaker/writer. Some oth-er rules define more specific cases.

The descriptive approach was more lenient to living English. Thisapproach consisted in collecting samples of living English language anddescribing the distribution (frequency) of structures, which were regularlyused in communication. Descriptive linguistics was the continuation of the

9

behaviourist tradition in language studies, focusing on the "observed be-haviour", "response to stimuli", and "repertoire of responses".

The descriptive approach gave rise to structural linguistics that setitself the goal to sort out the language material obtained through descrip-tion. It pictured language as a system with a highly organised structure ofdistinct meaningful elements. Meaningful elements of one language werethose that differentiated the meaning. E.g. the phonemes [p] and [b] aredistinctive in English but not in Arabic, where Pompeii and Bombay areunderstood as having the same initial phoneme. The goal of structurallinguistics was to describe the inventory of essential English structures.This resulted in a fairly symmetrical grammar system with the structuresclearly differentiating the meaning in communication.

The logic of descriptive studies is mostly inductive, proceeding fromthe numerous oral and written texts and ending up with some generalitieson how the language works. Descriptive grammar is used to answer thequestion "What is the language like?" unlike the prescriptive grammarwith its search for "What should the language be like?" E.g. prescriptivegrammar will consider which road sign is correct, "Drive slow!" or "Driveslowly!", while descriptive grammar most probably accepts both as alanguage reality. Other disputable cases, to name but a few, are "at theweekend" vs. "on the weekend" or "over the weekend".

Modern grammar tends to describe rather than to prescribe and sodo the dictionaries. When judging about the grammaticality of phrases"They don't have none" and "They don't have any", descriptive grammar-ians would observe both forms in common use therefore legitimizingboth. The same is true for "you and me" and "you and I", though as in theprevious case, some appear in spoken performance, while others in theedited published language.

1.2. TRADITIONAL GRAMMAR

1.2.1. Traditional Grammar in Ancient GreeceIn European tradition, the beginning of linguistics as a purposeful

and systematic study of language is ascribed to Ancient Greece. Thoughthe primary linguistic teaching in Ancient Greece was in many respectsnaive and speculative, and the native form of speech was the only formstudied, already in those times, from the 5th century B.C., fundamentalproblems were put forward that ran through ages of analytic linguisticeffort up to our days. Formally, traditional grammar is the type of gram-mar as it was before the advent of structural linguistics. Two periods of

10

traditional grammar could be distinguished: 1) prescriptive (pre-scientif-ic) and 2) descriptive (scientific).

Traditional grammar has its origins in the principles formulated bythe scholars of Ancient Greece and Rome – in the works of DionysiusThrux, Protagoras, Plato, Aristotle, Varro, and Priscian. Dionysius Thrux(c. 100 B. C.) was the first to present a comprehensive grammar of Greek.His grammar remained a standard work for thirteen centuries. Thrux dis-tinguishes two basic units of description – the sentence (logos), which isthe upper limit of grammatical description, and the word, which is theminimal unit of grammatical description. The sentence is defined no-tionally as "expressing a complete thought". The constituents of the sen-tence were called meros logos, i.e. parts of the sentence.

In their study of language, the Ancient Greeks considered the fourmain ranges of questions:

1. The most general, philosophical questions of language, such asthe origin of human speech.

2. Questions concerning structural categories in language, includ-ing phonetics.

3. Questions concerning usage, i.e. selection of words and construc-tions from the point of view of their 'correctness'.

4. Linguistic questions connected with the study of literary formsand rhetoric.

The central philosophical problem of language in Ancient Greecewas the problem of the relation between the words and the things theysignify. In connection with the controversy philosophers expressed theirviews about the nature and origin of human speech.

The two philosophers are commonly named as the main figures atthe outset of the dispute: Heraclitus (ab. 544 – ab. 483 B. C.) and Dem-ocritus (ab. 460 – 370 B. C).

According to Heraclitus and his followers, there is a natural con-nection between words and the things they signify: the nature of thingspredetermines the form of their names. Hence, language is inherent innature, and is given to people by nature. This conception of languagewas called 'fusei' – 'by nature'. It was idealistic. According to the greatmateri-alist philosopher Democritus and his followers, the connectionbetween words and the things they signify is the result of human conven-tion. Hence, language was created by the people themselves. This con-ception of language was called 'thesei – 'by convention', 'by law'.

A broad picture of the philosophical linguistic views prevalent inhis time was shown by Plato (427 – 347 B. C.) in his famous dialogueCratylus, or about the Correctness of Names. There are three personagesin the dialogue: Cratylus, Hermogenes, and Socrates. Their historical

11

prototypes are prominent philosophers. Cratylus defends the doctrine 'bynature' ('fusei') in an argument with Hermogenes, an adherent of the doc-trine 'by law' ('thesei'). The arguers ask Socrates to settle their dispute.Socrates exposes the schematism in the two opposite views, at the sametime finding grains of truth in both of them. But he points out the third,and the final, factor among those determining the meaning of the words,namely – usage in people's community.

Apparently in the discourse of Socrates the author's views were ex-pounded. The idea of usage as a factor determining the meaning of the wordwas one of the most profound linguistic conceptions formulated in AncientGreece. It was further developed both by philosophers and grammarians.

The criticism of the one-sided approach to the connection betweenthe 'linguistic sign' and the thing is revived in our time by some represen-tatives of modern linguistics.

In another dialogue, The Sophist, Plato analyses the nature of thesentence (or 'speech') as different from separate words. Here we may seethe rudimentary conception of the vocabulary as a set of naming ele-ments and the sentence as the result of connecting words in the processof speaking: the sentence is uttered for the sake of expressing the relationof the things to actual life. The minimum sentence, according to Plato,consists of a name plus a verb.

The first explicit grammatical teaching was propounded by Plato'sdisciple, the great Greek philosopher Aristotle (384 – 322 B. C.). Aristo-tle developed the theory of the sentence and the theory of word classes asnotional and functional parts of speech. But, being the founder of logic,Aristotle identified the relation of ideas in human thought with the rela-tion of words in speech, and stated grammatical categories in terms oflogic. He introduced in grammar the logical notions of subject and pred-icate. His criterion for discriminating between parts of speech was theability of words to express the parts of the logical proposition, i.e. thesubject, the predicate, and the copula. Accordingly, he established threeparts of speech: the 'name' and the 'verb' (forms expressing both the sub-ject and the predicate), and the 'conjunction' (forms expressing copulas).Thus, by 'names' he understood, in modern terminology, the nominativecase of nouns, adjectives, participles; by 'verbs', the infinitive of verbs;by 'conjunctions', different functional words and forms.

Proceeding from this fundamental thesis, he formulated the conceptof grammatical 'falls' ('cases') as deviations from 'names' or 'verbs' due tothe logically dependent position in the sentence, incapable of expressingeither the subject or the predicate. In the later grammatical tradition "thedoctrine of names" and their 'falls' was developed into the teaching of'direct' grammatical forms and 'oblique' grammatical forms.

12

Aristotle was the initiator of grammatical theory. But he lived along time before the final formation of grammar as a special discipline inAncient Greece. His original teaching was perfected and reformed by thelater scholars. Still, the confusion of categories of grammar with catego-ries of logic remained one of the main faults of the Greek grammarians.

The grammatical teaching of Ancient Greece was completed in Alex-andria, between the 2nd century B. C. and the 2nd century A. D. The devel-opment of grammar in Alexandria was stimulated by the interpretations ofHomer's poems that were extremely popular, but the language of whichhad become antiquated. Gradually grammar became a self-dependent dis-cipline taught and studied not by philosophers, but by grammarians.

The best-known scholars of that period were Aristarchus (ab. 217 –145 B. C.), Dionysius Thrux (ab. 170 – 90 B. C), and Apollonius Dysco-lus (the 2nd century A. D.).

Thrux distinguished onoma (noun) class words, rhema (verb), me-tochз (participle), arthron (article), antonymia (pronoun), prothesis (prep-osition), epirrhзma (adverb), and syndesmos (conjunction). He reunitedthe Stoic common and proper nouns into the single onoma (noun) class;he separated the participle from the verb. The adjective was classed withthe noun, as its morphology and syntax were similar to those of nouns.The noun was defined as a part of the sentence inflected for case andsignifying a person or a thing; the verb as a part of the sentence withoutcase inflexion, but inflected for tense, person, and number, signifying anactivity or process performed or undergone; the participle as a part of thesentence sharing the features of the verb and the noun; the article as apart of the sentence inflected for case and preposed or postposed to nouns;the pronoun as a part of the sentence substitutable for the noun and markedfor person; the preposition as a part of the sentence placed before otherwords; the adverb as a part of the sentence without inflexion, in modifi-cation of or in addition to the verb; the conjunction as a part of the sen-tence binding together the discourse and filling gaps in its interpretation.

Each defined class of words is followed by a statement of the catego-ries applicable to it. Thrux refers to them as parepomena. By parepomenahe means grammatically relevant differences in the forms of words whichinclude both inflexional and derivational categories. To illustrate this, con-sider the noun. Thrux distinguishes five such categories of the noun:

1) Genos (gender): masculine, feminine, neuter;2) Eоdos (type): primary or derived;3) Schзma (form): simple or compound;4) Arithmos (number): singular, dual, or plural;5) Ptфsis (case): nominative, vocative, objective, genitive, dative.The parepomena of the verb included mood, voice, type, form, num-

13

ber, person, tense, and conjugation. Three basic time references are distin-guished: present, past, and future. Thrux’s set of parts of speech has under-gone only minor modifications and is still very much in use today. Themain omission in this grammar is the absence of any section on syntax.

Syntax was dealt with, rather extensively, by Appolonius Dyscolus.Appolonius based his syntactic description on the relations of the nounand the verb to each other and of the remaining word classes to thesetwo. The achievements of the Greek scholars lie in devising and system-atizing a formal terminology for the description of the classical Greeklanguage, a terminology which, through adaptation to Latin and later onadopted from Latin by other languages, has become part and parcel ofthe grammatical equipment of the linguistics of our day.

1.2.2. Traditional Grammar in Ancient RomeRoman linguistics was largely the application of Greek thought to

the Latin language. The relatively similar basic structures of the two lan-guages facilitated the process of this metalinguistic transfer.

The first Latin grammar was written by Varro (116 – 27 B. C.). HisDe Lingua Latina comprised 25 volumes. One of Varro’s merits is the dis-tinction between derivation and inflexion. Inflexional formations are char-acterized by great generality; they do not vary in use and acceptabilityfrom person to person and from one word root to another. The former partof morphology Varro called declinatio naturalis (natural word variation)and the latter, declinatio voluntaria (spontaneous word form variation).Varro set up the following system of four inflexionally contrasting classes:

1) those with case inflexion (nouns including adjectives);2) those with tense inflexion (verbs);3) those with case and tense inflexion (participles);4) those with neither (adverb).The Latin grammars of the present day are the direct descendants of the

works written by late grammarians, Priscian (c. A. D. 500) in particular. Hisaim, like theirs, was to transfer as far as he could the grammatical system ofThrux’s grammar, as well as the writings of Appolonius, to Latin. He uses theclassical system of eight word classes laid down by Thrux and Appolonius,with the omission of the article and the inclusion of the interjection. Pris-cian’s work is based on the language of the best writers (e.g. Cicero, Virgil),i.e. not on the language of his own day. Priscian’s work marks the bridgebetween Antiquity and the Middle Ages in linguistic scholarship.

In the works of the Alexandrian scholars many features of grammarwere shaped into the form that the linguists of the 19th century called'traditional' grammar. Aristotle's doctrine of 'names' and their 'cases' was

14

reformed. The words of the language were grouped into eight parts ofspeech: inflected (name, verb, participle, article, pronoun) and uninfected(preposition, adverb, conjunction). The chief grammatical categories('aceidenies') of the inflected parts of speech were described, such as gen-ders, numbers, cases (the categories of the nominal parts of speech); num-bers, persons, tenses, moods, voices (the categories of the verbs). The sen-tence and the word were considered the chief elements of connected speech.The sentence was defined as "a combination of words expressing a com-plete thought" (Dionysius Thrux). The word was defined as "an articulatesound with a certain meaning out of which the sentence is composed, andinto which it is decomposed" (Diomedes, the Roman grammarian of the 4th

century A. D., a follower of the Alexandrian school).Some progress was made in studying the phonetical structure of

speech. Speech sounds ('letters') were classified into vowels, semi-vow-els, and non-vowels, the main principle of the classification being thesyllable-forming function of the sounds. The Alexandrian scholars de-scribed the word accents, some phonetical changes in the process of speak-ing, the difference between long and short vowels. Observations of thephonetical phenomena were made chiefly in connection with the studyof metrics and prosody.

As for general considerations of language, the argument about anal-ogy and anomaly took place among the Alexandrian scholars, that con-tinued during many centuries and was inherited by Rome. The 'anomal-ists' taught that language does not correspond to our concepts of things,and therefore it is il-logical and irregular. This doctrine was first pro-pounded by the stoics (3rd-2nd centuries B. C.) The 'analogists' argued thatlanguage, on the contrary, does correspond to our ideas of things, and itis quite logical and regular.

It is easy to see in this discussion traces of the philosophical disputeabout the correctness of names. But the Alexandrian scholars, in particu-lar Aristarchus, who argued with the anomalists and began the discus-sion in Alexandria, soon turned it to more concrete, grammatical lines.The argument served as an incentive to closer study of language. Com-pare the thesis of the anomalists that the only criterion for correctness inspeech is the popular usage, and not 'the art of grammarians' (SextusEmpiricus, 2nd century A. D.).

The weakest point of language study in those times, both with theGreek and the Roman scholars, was the problem of etymology. The ety-mologies given were absolutely fantastic. The division of words intomeaningful component parts was quite arbitrary. Likewise the morpho-logical composition of the word remained alien both to the Greek and tothe Roman grammarians.

15

The Romans who were successors to the culture of Ancient Greece,succeded also to the Greek linguistic theories and grammatical teaching.Moreover, our knowledge of many lin-guistic achievements of the Greeksis derived only from Roman sources.

In particular it refers to the argument about analogy and anomaly, whichis known mainly from the treatise of Varro (116 – 27 B. C), one of the great-est Roman grammarians. Varro himself contributed many considerations tothe argu-ment, coming to the conclusion that in the sphere of Latin inflectedforms analogy prevailed, while word-building was more subject to anomaly.

On the whole, the Romans did not care as much as the Greeks aboutgeneral linguistic problems. They constructed their grammar on the Greekmodel, with some modifications. Due to the fact that it had been preced-ed by the Greek grammatical teaching, their grammar seems to be moresystematic and completed from the outset. Much attention was paid inRome to the problems of style and rhetoric, in particular to the problemof stylistic types of speech. The most prominent figure in this field wasM. T. Cicero (106 – 43 B. C.), the great Roman orator.

The highest authority among the Roman grammarians belongs toDonatus (4th century A. D.) and Priscian (6th century A. D.). The twoscholars generalized the grammatical observations of Latin made by theirpredecessors during several centuries. In their comprehensive works thegrammatical teaching of Antiquity found its most complete expression.

In the Middle Ages Latin became the universal language of Catho-lic Church, and also of scholastic science. The grammars of Donatus andPriscian (especially the former) and adaptations of them remained in usethroughout centuries up to the epoch of the Renaissance.

As regards science, the Middle Ages are characterized as a time ofstagnation, due to the domination of Church. Linguistic theory was, in themain, confined to biblical dogmas. The origin of language was ascribed todivine creation. Likewise, the fact of the diversity of languages on theearth was explained by means of the biblical legend about the Tower ofBabel: God confused the languages of the peoples to prevent them fromreaching heaven. Still, even at that time scholars made some further impor-tant observations about Latin grammar. They defined nouns and adjectivesas different parts of speech within the class of names, and also discoveredsyntactical categories of concord, government, and apposition.

But the scholastic philosophers of the Middle Ages saw in the struc-ture of Latin the only natural and logically perfect form of speech ingeneral. They proceeded from the assumption that all the languages hadessentially the same basic structure as Latin, and presupposed the same'logical normal' grammatical categories in all the languages of 'lesserimportance' than Latin, i.e. the European languages of their time.

"

16

The conception that the structure of different languages is based onthe same logical, rational categories was developed further throughoutthe epoch of the Renaissance (14 – 16th centuries) and in the 17th centuryled to the theory of universal grammar based on logical principles, withthe same fundamental categories for all languages.

The first and most famous linguistic work inspired by this theorywas written by Lancelot and Arno at the convent of Port Royal in theyear 1660. It was entitled «Grammaire Generate et Raisonnee» – Gener-al and Rational Grammar. It demonstrated the would-be identity of gram-matical categories in all the languages on the example of French, Latin,and Greek. Many other grammars appeared in different European coun-tries taking the grammar of Port Royal as a model.

The theory of logical grammar was subjected to severe criticism byrepresentatives of modern linguistics. It has been stressed that scientificlinguistics must form the system of general linguistic categories as a gen-eralization of the study of various languages, and not as pre-conceivedformulas. Still, other scientists, especially the propounders of transfor-mational grammar undertake to revise this criticism. They point out that,though the theory of rational grammar was limited by the conditions andviews of its time, its basic principles were fruitful and sound; above all,its trend was not only to describe linguistic facts, but to explain themalong 'generative' lines, showing how more complex structures of speechare created out of elementary structures.

But the epoch of the Renaissance brought to the field of linguisticsnot only the idea of general rational grammar. The horizon of linguisticswidened. New European languages came to be described in grammarbooks. The study of Ancient Greek was resumed in Europe, and somescholars began to study Hebrew and Arabic. As a result of geographicaldiscoveries, European scholars got acquainted with the languages of 'ex-otic' countries. In the 16-17th centuries many grammars and dictionarieswere compiled of languages that had been unknown before.

This work was crowned towards the end of the 18th century by compil-ing polyglot glossaries that contained parallel lists of words translated intoas many languages as the edi-tors could gain information of. The summaris-ing work in this field was published in Russia by the Academy of Sciencesunder the editorship of P. S. Pallas, in the years 1786 – 1787. It was entitled«Cpaвнительные словари всех языков и наречий» and included 286 wordsgiven in more than 200 languages (the second edition published in 1791,included 272 languages). In Germany, in the years 1806 – 1817, J. C. Ade-lung and J. S. Vater published a huge treatise containing translations of theLord's Prayer into 500 languages with glossaries and commentary. The trea-tise was entitled Mithridates, or General Linguistics.

17

Alongside of this work, from the epoch of the Renaissance philolo-gists began to collect and publish historical monuments of languagesother than Latin or Greek. In this way materials for the future scientifichistory of languages were prepared.

The scope of all this language study was tremendous. But the scien-tific foundation of the work was inadequate. That which might be con-sidered as achievements of linguistics in the ancient world, cannot beconsidered as such for the period in question. Scholars continued to statemany linguistic categories in terms of logic, taking no notice of the struc-tural difference between languages. They forced the description of dif-ferent languages into the traditional scheme of Latin grammar. They hadno proper understanding of the difference between sounds and letters.They did not understand the nature of local dialects, considering them tobe the 'corruption' of the 'correct' form of language, i.e. of the writtenlanguage of books, or the speech forms used in 'refined' society. Theyhad no idea of historical language development. Unscientific views wereexpressed about the origin of language.

In spite of these drawbacks, however, the linguistic work done fromthe time of the Renaissance was of great importance. It demonstrated theactual diversity and multitude of languages on the earth. It disclosed thefact that all the living languages were equally effective as means of so-cial intercourse. Due to the study of different languages materials weregradually collected that were necessary for the creation of modern, sci-entific linguistics.

1.2.3. Traditional Grammar in Ancient IndiaOutside the tradition of European linguistics, considerable progress

in the study of language was made in Ancient India. The Ancient Hinduscholars are not so highly distinguished for raising general linguistic prob-lems as the scholars in Ancient Greece; but the grammar that they creat-ed is in many respects a higher achievement than the grammar of theAncient Greeks: it gives a much more rigorous and objective descriptionof the structural elements of language.

The factors that gave rise to grammatical teaching in Ancient Indiawere connected in a peculiar way with the interpretations of the old reli-gious texts – the Vedas, large collections of hymns, songs, ritual formu-las, and incantations (the Old Hindu 'Veda' means 'knowledge'). The old-est of the Vedas, the Rig-Veda ('the knowledge of hymns') dates from thethird – second millennium before our era. Many centuries before theyhad been committed to writing, the Vedic texts were handed down fromgeneration to generation by an oral tradition. The oral teaching of them

18

was extremely exact, because they were to be preserved unchanged fromreligious motives. Meanwhile, the living language developed and be-came more and more different from the old Vedic language. Commentar-ies appeared interpreting the Vedic passages the meaning of which hadbeen lost or obscured. New works were also added to the texts, the lan-guage of the whole Vedic literature becoming more and more heteroge-neous as the time went by.

Somewhere about the 6th century B. C. scholars began to normalisethe language of the various late additions to the original texts, taking astheir model the form of speech used by the Brahmins, scholars, court po-ets. In this way Sanskrit appeared – the 'perfected', 'learned' language ofthe upper caste ('samskrta’ in Old Hindu means 'perfected','elaborated').The common language of the people, as opposed to Sanskrit, was termed'Prakrit' by later scholars ('prakrta' in Old Hindu means 'plain', 'popular').The work of describing and normalising Sanskrit was completed in the 4thcentury B. C. by the famous Hindu scholar Panini, who gave a detaileddescription of Sanskrit and, partly, of the Vedic language. In the 2nd centuryB. C. Panini's grammar was supplemented with an important commentaryby another prominent grammarian, Patanjali. The grammar of Panini andPatanjali's commentary are the oldest and the best systematic grammaticalworks of the Ancient Indians that have come down to us.

Panini canonised the forms of Sanskrit as the 'sacred' language, thelanguage of religious worship. But later the use of Sanskrit went beyondthese limits, and great secular literature was created in it. The languageof this period is called 'Classical Sanskrit', to distinguish it from the Vediclanguage, or the 'Vedic Sanskrit', as it was called by the European lin-guists. Classical Sanskrit was used in India throughout the Middle Ages,similar to Latin in Europe.

Panini's grammar is the result of colossal linguistic effort that couldnot have been achieved by one man. Panini himself refers to a number ofhis predecessors, though their works have not been preserved.

The treatise is entitled 'Astadhyayl' – Eight Readings; it consists ofeight books, or chapters, containing about 4000 very short grammaticalrules – 'sutras', given in verses. The 'sutra' form of the text is connectedwith the fact that it was meant for learning by heart, and chiefly fromhearing. Symbolic notation is used in the work, to make the memorisingeasier. For example, special letters are used to signal the posi-tions of thestress in the words etc.

Sanskrit was a language with strongly pronounced synthetic features.It was exceptionally rich in inflexions, widely used vowel interchange asgrammatical means, and had an extremely developed system of word-build-ing. Panini's description adequately answered these features.

19

One of the main achievements of the Old Hindu grammatical theo-ry was that it discovered the morphological structure of the word: theroot, the stem, the suffix. It gave a detailed description of the phoneticalform of the root, discovering the different grades of vowel interchange.The words of the language were classified according to formative char-acteristics, the primary verb roots being considered as the basic sourcefor all the vocabulary. All the types of declination and conjugation wereinvestigated. Syntactical study was also well advanced, being partly com-bined with the study of word composition, which was due to the impor-tant role of compound words in the Sanskrit sentence.

The phonetical description given in the Old Hindu grammar is es-pecially accurate; it is connected with the purpose of the work: to showhow to pronounce the sacred texts without any distortions. The organs ofspeech were studied carefully; the sounds were described in accord withtheir articulation, much attention being given to the active organs ofspeech: the lips, the three parts of the tongue (front, middle, and back),the larynx. Phonetical changes on the borders of words and affixes werealso analysed with precision.

Alongside of Sanskrit forms, Panini's grammar described certainforms of the Vedic language, thereby containing elements of compara-tive language study.

Some knowledge of Sanskrit and of Panini's grammar reached Eu-rope in the 16th and 17th centuries. Towards the end of the 18th century itwas studied diligently by European scholars-orientalists. At the begin-ning of the 19th century Panini's grammatical treatise Eight Readings waspublished in Europe.

The Hindu grammar of Panini presented to the European scholarsan accurate description of a language based not upon abstract specula-tion, but upon careful and exact observation. The Hindu grammar helpedto formulate one of the most important principles of scientific linguis-tics: to study the constituent parts of a language without any predeter-mined conclusions. This consequence of the knowledge of Panini's gram-mar was effected in full later on, at the end of the 20th century, whensome languages of the dying out American Indian tribes came to be stud-ied, without any previous knowledge of them (see further Ch. Six). Itwas not accidental that Leonard Bloomfield, the father of American De-scriptive Linguistics, called Panini's grammar 'one of the greatest monu-ments of human intelligence'.

Apart from this, however, the knowledge of Sanskrit and the Hindugrammar led to a discovery of tremendous significance. The European scholarssaw that Sanskrit had a structure very similar to the structure of Latin, Greek,and some other European languages, both old and new. The first reaction to

20

this fact was the idea that Sanskrit was the source from which all the Europeanlanguages had sprung. But this view was later rejected, and it was understoodthat Sanskrit, being related to Latin and Greek, together with them formedpart of a great family of kindred languages.

This discovery, made at the end of the 18th century, had arevolutionising effect on linguistics, marking the turning point in itsdevelopment. The comparative study of Sanskrit and European languagesgave rise to the historical comparative linguistics, and led to thecompletion of linguistics as a science in the full sense of the word.

Check yourself test 1Part 1

1. In what European country did the study of language begin?What was the central philosophical problem of language in Ancient

Greece?What two philosophers are named as the main figures in the "dis-

pute about the correctness of names"?How did they explain the connection between the words and the

things they signify?What doctrine was called 'by nature'?What doctrine was called 'by law'?Who described the philosophical linguistic views prevalent in his

time in Ancient Greece in the 4th century B. C.?What linguistic views did Plato express in his works?2. What grammatical categories were established by Aristotle?What was the relation between logical and grammatical categories

in Aristotle's system?What parts of speech did he discriminate?What forms did he call 'names' and what forms did he call 'cases'?3. How was grammar developed in Ancient Greece after Aristotle?Where was the grammatical teaching of Ancient Greece completed?Who were the most famous grammarians of the Alexandrian period?What grammatical observations did they make? What was their

progress in the study of phonetics? What linguistic dispute took placeamong the Alexandrian scholars?

What were the weakest points of language study in Ancient Greece?4. How did linguistics progress in Ancient Rome?Who were the most prominent Roman grammarians?How did the Romans construct their grammatical teaching?What were their achievements in language study?

21

5. How was language study developed in the Middle Ages?Why are the Middle Ages characterised as a time of stagnation for

science?What language became the universal language of Catholic Church

and the scholastic mediaeval science?What new grammatical observations of Latin were made in those times?How was the structure of languages understood by the scholastic

philosophers?6. What were the achievements of the study of language from theepoch of the Renaissance up to the 18th century?What do you know about the grammar of Port Royal?What languages came to be studied at the epoch of the Renaissance?What summarising polyglot works were published towards the end

of the period?What is the significance of the linguistic work done during that pe-

riod?What were the main drawbacks of this linguistic work?

Part 2 1. How did grammatical teaching appear in Ancient India?What old texts needed interpretation in Ancient India? Why?What is Sanskrit?How did it originate?How did it develop later?Who gave the oldest grammatical description of Sanskrit?2. What are the main characteristic features of Panini's grammar?What accounts for the 'sutra' form of the rules given in the work?What important grammatical observations did Panini make in his

grammar?How was the phonetical structure of Sanskrit described?Why was so much attention given to phonetics in the Hindu gram-

mar of Sanskrit?Why do many scientists consider the Hindu grammar of Sanskrit as

the highest linguistic achievement of ancient times?3. How did the knowledge of Sanskrit and its grammar influence

European linguistics?When did the Europeans get acquainted with Sanskrit and with Pa-

nini's grammar?In connection with what studies did Panini's grammar influence

modern linguistics?What discovery was made due to the comparison of Sanskrit with

some European languages?What was the consequence of this discovery?

22

1.3. PRESCRIPTIVE GRAMMAR

As already known, until the end of the sixteenth century, the onlygrammars used in English schools were Latin grammars. The aim was toteach Englishmen to read, write and sometimes converse in this linguafranca of Western Europe.

One of the earliest and most popular Latin grammars written inEnglish was William Lily’s grammar, published in the first half of the 16th

century. It was an aid to learning Latin, and it rigorously followed Latinmodels.

The Renaissance saw the birth of the modern world. It widenedlinguistic horizons. Scholars turned their attention to the living languag-es of Europe. Although the study of Greek and Latin grammar continued,they were not the only languages scholars were interested in. As can beexpected, the first grammars of English were closely related to Latingrammars. Latin had been used in England for centuries, scholars hadtreated it as an ideal language. They were struck by its rigor and order.English, which replaced Latin, had to appear as perfect as Latin. As aresult, some English scholars were greatly concerned with refining theirlanguage. Through the use of logic they hoped to improve English.

The first grammars of English were prescriptive, not descriptive.The most influential grammar of this period was R. Lowth’s Short Intro-duction to English Grammar (1762). The aim of this grammar was "toteach us to express ourselves with propriety... and to enable us to judgeof every phrase and form of construction, whether it be right or not".Unfortunately, the criterion for the discrimination between right and wrongconstructions was Latin. As Latin appeared to conform best to their con-cept of ideal grammar, they described English in terms of Latin formsand imposed the same grammatical constraints. For instance, a noun waspresented in the form of the Latin noun paradigm:

Nominative: the houseGenitive: of the houseDative: to the houseAccusative: the houseAblative: in, at, from the houseVocative: housePrescriptivists promoted those grammatical variants which corre-

sponded, in one way or another, to equivalents in Latin. Anxious to do it,they prescribed and proscribed many of the constructions used in En-glish from time immemorial. They condemned the use of a preposition insentence-final position, e.g. who are you looking at? or who are you talk-ing to? The reason for the condemnation was that sentences do not end

23

with a preposition in Latin. But even in Old English we could find sen-tences ending with prepositions. The rule "It is incorrect to end a sen-tence with a preposition" was repeated in prestigious grammars towardsthe end of the eighteenth century, and from the nineteenth century on itwas widely taught in schools.

Another restriction that the prescriptivists applied to English wasthe Latin constraint on the use of the accusative form of a noun after theverb esse (to be). Since me is historically the accusative form of the per-son (nom.: I; gen.: my; dat.: to me; acc.: me; abl.: by me; voc.: of me), itwas considered wrong to say it’s me. Instead we must say It’s I. Thepattern It’s me, which had been common for centuries and still is, wasthought incorrect since the Latin construction ego sum made use of thenominative form of the pronoun.

Another prescription was not to use the construction better thanhim. Writers of Lowth’s era used both better than he and better than him.His preference for the former he explained by the fact that better than hecan be followed by the verb is and better than him cannot.

Prescriptivists disliked variation and change. Correctness was asso-ciated with what used to be the case. Different from was preferable todifferent to, or different than, because the di-part of the word in Latinindicated division or separateness, and therefore from suits the etymo-logical argument better.

Prescriptivists condemned constructions on account of logic as well.For instance, had rather and had better, double comparatives (lesser,worser) were regarded as contradicting the laws of reason. Logic wasused to stigmatise some constructions and promote others. The most no-torious example concerns double negation, e.g. I don’t know nothing.Such patterns were traditional. Shakespeare used double negation. How-ever, they were condemned as incorrect. Last but not least, prescriptiv-ists disregarded English of their day: they would rather draw their exam-ples from the past. Even the English of the best writers of the past wassometimes regarded as wrong if it did not correspond to their conceptionof correct English.

Prescriptivists are conservative linguists: when there is a competitionbetween an older form and a more recent alternative, they dislike changewhich is identified with corruption: the language of their ancestors hadbeauty, but the language of his contemporares is always diminished.

Latinization of English grammar was also reflected in the system ofparts of speech. Patterning after Latin, prescriptivists set up a classifica-tion of eight parts of speech: noun, pronoun, verb, adjective, adverb,preposition, conjunction, and interjection. The English articles a(n) andthe, having no Latin counterparts, were not given the status of a part of

24

speech, but merely referred to as signs before nouns to identify them asnouns. Some prescriptivists treated the articles as a subclass of adjec-tives. Only Ben Jonson assigned them to a class of their own. Similar toLatin grammarians, prescriptivists, in defining word classes , syntacticstructures, relied either on meaning or function.

To sum up, prescriptive grammar could be characterized by the fol-lowing features:

1) Patterning after Latin in classifying words into word classes andestablishing grammatical categories;

2) Reliance on meaning and function in definitions;3) Approach to correctness: the standards of correctness are logic,

which was identified with Latin, and the past.4) Emphasis on writing rather than speech.As prescriptive grammarians were concerned with the rules for the

correct use of English, they could be called the first standardizers ofEnglish. Unfortunately, their "standardization" work was often based onsubjective criteria and other languages. However, not all works writtenin the prescriptive era ignored actual usage. Those which did not pavedthe way to Standard English, which has today become an objective stan-dard for correct English. Those grammarians who adhere to the norms ofStandard English (the English of government, education, broadcasting,news publishing, and other public discourse) are also prescriptivists –prescriptivists in a good sense.

1.4. THE HISTORICAL COMPARATIVE METHODIN LINGUISTICS

There are many languages on the earth, both great and small.According to modern calculations, the number of living languages exceeds2500. Alongside of highly developed national languages with ancientwriting and literature, there are languages having no writing and norecorded history: here belong the spoken languages of tribes and smallnationalities in America, Asia, Africa, Australia. Many of the spokenlanguages are dying out together with their peoplesdue to the miserablecondition they have been reduced to by the "higher European civilisation",as is the case with the aboriginal Indian tribes in America or Australia.On the other hand, the number of known languages is still growing, asnew languages and dialects come to be recorded and studied by science.

Observing the fact that some of the languages are very similar toone another in their forms, while others are quite dissimilar, scholars stilllong ago expressed the idea that languages revealing formal features of

25

similarity have a common origin. Attempts to establish groups of kindredlanguages were repeatedly made from the 16th century on. Among thescholars who developed the idea of language relationship and attemptedto give the first schemes of their genealogical groupings we find the namesof J.J.Scaliger (1540 – 1609), the great German scientist G. W. Leibnitz(1646 – 1716), the great Russian scientist M.V.Lomonosov (1711 – 1765),and others.

But a consistently scientific proof and study of the actual relationshipbetween languages became possible only when the historical comparativemethod of language study was created in the first quarter of the 19thcentury.

The historical comparative method developed in connection withthe comparative observation of languages belonging to the Indo-Europeanfamily, and its appearance was stimulated by the discovery of Sanskrit.

Sir William Jones (1746 – 1794), a prominent British orientalist andSanskrit scholar, was the first to point out the form of rigorously groundedscientific hypothesis that Sanskrit, Greek, Latin, Gothic, and some otherlanguages of India and Europe had sprung from the same source which nolonger existed. He put forward this hypothesis in his famous report to theCalcutta Linguistic Society (1786), basing his views on an observation ofverbal roots and certain gram-matical forms in the languages compared.

The relations between the languages of the Indo-European familywere studied systematically and scientifically at the beginning of the 19th

century by Franz Bopp (1791 – 1867), Rasmus Kristian Rask (1787 –1832), Jacob Grimm (1785 – 1863), and A.Ch.Vostokov (1781 – 1864).These scientists not only made comparative and historical observationsof the kindred languages, but they defined the fundamental conceptionof linguistic 'kinship' ('relationship'), and created the historical comparativemethod in linguistics. The rise of this method marks the appearance oflinguistics as a science in the strict sense of the word.

After that the historical and comparative study of the Indo-Europeanlanguages became the principal line of European linguistics for many years.

The historical comparative linguistics was further developed in theworks of such scientists of the 19th and 20th centuries as F.Dietz (1794 –1876), A.F.Pott (1802 – 1887), A.Schleicher (1821 – 1868), F.I.Buslayev(1848 – 1897), F.F.Fortunatov (1848 – 1914), F.de-Saussure (1857 – 1913),A.Meillet (1866 – 1936) and other linguists.

The historical comparative method is a system of analytical proce-dures applied to the study of languages in their historical development.The historical comparative method is used to analyse and discover therelationship of different languages and groups of languages, to recon-struct prehistoric lingual elements (prehistoric in the sense that they are

26

not fixed in written monuments), to reveal the course of historical devel-opment of lingual elements in their complex interrelations; by means ofthe historical comparative method science collects materials for study-ing general laws of language development.

The following general conceptions of different aspects of languageand its development underlie the foundations of the historical compara-tive method: (1) families of languages originate due to historical divisionof languages; (2) lingual signs (signals) are arbitrary in the sense thatthere is no natural connection between their form and the things or ideasthey signify; (3) the historical development of language is continual, butuneven. We shall consider these fundamental conceptions and their con-sequences separately.

1. The historical comparative method proceeds from the possibilityfor different languages to have been originated from the same source.The division of one language into two or more languages is brought aboutby the division of the language-speaking community due to political andeconomic factors. Since language is always changing historically, theisolation (full or partial) of daughter communities can lead to the grow-ing differences in their language, to the rise of dialects, which, in theprocess of further change, can develop into totally different (though re-lated) languages.

Such division, or 'differentiation', of languages is characteristic ofthe tribal epoch in the history of peoples. The actual process of languagedivision is very complex. It is connected with repeated mixings, cross-ings, and re-divisions of tribes and nationalities throughout centuries andmillenniums; and accompanied by the disappearance of some languages,and the spread of other languages over vast territories and among origi-nally unrelated communities. One thing, though, is absolutely straight-forward: when the dialects of a language grow into different languages,it means that the parent language has ceased to exist: in its stead a familyof languages has arisen. Thus, in the family forming linguistic processwe register several stages of differentiation corresponding to the exist-ence of several successive parent languages: the parent language of afamily and the intermediary parent languages of further groups and sub-groups within the family. E.g. in the genealogical prehistory of the Indo-European family we distinguish the Parent ('Primitive') Indo-Europeanlanguage, and its direct and indirect descendants: the Parent ('Primitive')Indo-Iranian, the Parent ('Primitive') Balto-Slavonic, the Parent ('Primi-tive') Germanic, etc. This idea of language relationship was alien to schol-ars before the rise of scientific linguistics.

27

2. The actual kinship (relationship), or non-kinship of dif-ferent lan-guages (i.e. their being the further forms of the same underlying parentlanguage) is revealed on the basis of systematic comparison of their forms:since there is no naturally predetermined connection between lingualforms and the things or ideas they signify, the resemblance of meaning-ful forms in different languages can evidence their kinship.

Different cases of resemblance must be considered here.Occasionally resemblance in meaningful forms is purely coinciden-

tal. E.g. the Malay [mata] meaning 'eye' closely resembles the ModernGreek [mati ] with the same meaning, but the resemblance is due to coin-cidence, which is proved by the history of the two languages. Likewisethe Georgian conjunction [da] meaning 'and' is similar to the Russianconjunction 'Aa' by mere coincidence. Such cases of resemblance areextremely rare.

Other resemblances may be due to the borrowing of words fromone language to another, and therefore bear no evidence as to the actualkinship of the languages. E.g. the Georgian [aritmetika ] is similar inform and in meaning to the Russian 'apифметика, the English 'arith-metic', the French 'anthmetique', etc., by virtue of the fact that the word isborrowed from Ancient Greek (Gr. 'arithmetike').

However, there are such features of resemblance between languag-es that clearly prove their common origin. These features belong to themost stable component parts of language – to the basic word stock and,above all, to the fund ot grammatical affixes, because grammatical forms,as a rule, are never borrowed by one language from another.

By comparing forms in kindred languages, linguists reveal the sys-tem of phonetical correspondences characterising one language or groupof languages within the family in reference to another language or groupof languages.

For example, the Indo-European [p], [t], [k] correspond to the Ger-manic [f], [p], [x]; the Indo-European [b], [d], [g] correspond to the Ger-manic [p], [t], [k]; the Indo-European [bh ], [dh ], [gh ] correspond to theGermanic [b ], [d], [g] (the Germanic consonant shift discovered byJ.Grimm). Further, the Germanic [d] corresponds to the German [t], theGermanic [p] corresponds to the German [d], the Germanic [t] correspondsto the German [s] (after a vowel) or [z] (in other positions), etc. (the secondGermanic consonant shift). Such systems of correspondences are the finalsign of language kinship, they show both differentiation and unity.

3. Language develops unevenly. It concerns all the structural elementsof language, and also different languages as compared to each other on thesame chronological level. It is connected with the fact that different struc-

28

tural elements of language specifically react to, and reflect, the history ofthe people. It follows from this that elements no longer existing in onelanguage of a family may be preserved in other kindred language. Thus,comparing different languages and their forms linguists can reconstructprehistoric elements of languages, and more rigorously formulate the his-torical changes in languages (known from written monuments).

Reconstructing the prototypes, linguists gradually recreate the ba-sic phonetical, grammatical, and lexical features characterising the par-ent language of a family.

The historical comparative method has certain limitations.1. It is limited by the material it can use. The facts that ceased to

exist in all the compared group or family of languages can hardly bereconstructed.

2. Some common features of living kindred languages can be the prod-uct of further development; but in studying the languages comparatively andelaborating reconstructions there is a great danger of considering them ascharacteristics of the parent language, thereby distorting history.

3. It is difficult and sometimes impossible to define the time, andeven the relative chronology of lingual changes.

4. The historical comparative method can be chiefly ap-plied tolanguages with ancient writing, that is to languages having a long writtentradition or 'history'.

5. It is applied only to the comparative study of languages; but, tounderstand the innermost nature of language, all languages must be stud-ied in comparison, not only kindred. To bridge this gap, modern linguis-tics is developing the typological study of languages, both kindred andnon-kindred.

However, these limitations of the historical comparative methodcannot testify that the method is antiquated. The historical comparativemethod has itself undergone a long history. The historical comparativestudy of languages in the 19th century was still mixed with psychologicaland logical pre-conceived ideas. Towards the end of the 19th century itconcentrated its attention on the history of separate lingual elements,losing sight of their interrelations in the system of language. Such ap-proach to language was called 'atomistic'. Modern linguistics is gradual-ly overcoming this approach, perfecting its analytical and descriptivetechnique in historical studies. Alongside of the Indo-European familyof languages, other families of related languages have been discoveredand are being studied, such as the Finno-Ugrian family, the Turkish fam-ily, the Caucasian family, and others. Their study brings new scientificresults, widening the horizon of the comparative linguistics and contrib-uting to its development.

29

1.5. NON-STRUCTURAL DESCRIPTIVE GRAMMAR

In the second half of the 19th century the development of prescrip-tive grammar was completed. The best prescriptive grammars of the pe-riod, C.P.Mason’s English Grammar, 1858 and A.Bain’s Higher EnglishGrammar, 1863, paved the way for the appearance of a new type of gram-mar, viz. descriptive, or scientific grammar: a need was felt for a gram-mar which could give a scientific explanation of the actually occurringstructures without assessing the correctness of the structures.

Henry Sweet (1845 – 1912), the father of a new approach to linguis-tic studies, described it in the preface to his work, New English Gram-mar, Logical and Historical (1891) as follows: "As my exposition claimsto be scientific, I confine myself to the statement of facts, without at-tempting to settle the relative correctness of divergent usages. If an ‘un-grammatical’ expression such as it is me is in general use among educat-ed people, I accept it as such, simply adding that it is avoided in theliterary language" [28, p. 123].

Similar to prescriptive grammarians, Sweet mostly concerned him-self with the written language, the language of the best writers of histime. Sweet also adopted the grammatical system of his predecessors,but in classifying words into word classes he was more explicit as re-gards the criteria, or principles, of classification than prescriptivists.

The scholar seemed to adhere to the same conception of parts ofspeech as his ancient colleagues, viz. parts of speech are syntactic cate-gories – they manifest themselves in the sentence as relational catego-ries: the noun is related to the verb, the adverb is related to the verb, thepreposition is related to the noun, the adjective is related to the noun.

This approach can be clearly seen in Sweet’s description of the noun:"As regards their function in the sentence, words fall under certain class-es called parts of speech, all the members of each of these classes havingcertain formal characteristics in common which distinguish them fromthe members of the other classes. Each of these classes has a name of itsown – noun, adjective, verb, etc. If we examine the meanings of the wordsbelonging to the different parts of speech, we shall find that such nounsas tree, snow, man, are all substance-words" [28, p. 123].

The term scientific grammar means reliance on facts and the use ofthe inductive method.

Being interested in phonetics, Sweet could not ignore the spokenlanguage: The first requisite is a knowledge of phonetics of the form oflanguage. We must learn to regard language solely as consisting of groupsof sounds, independently of the written symbols. This is in fact the rec-ognition of the priority of oral speech over written [28, p. 143].

30

Henry Sweet was the first to undermine the old tradition in linguis-tic studies where the function of grammar was to prescribe what is judgedto be correct rather than describe actual usage. Among his followers wecan mention Poutsma, Kruisinga, Zandvoort, Curme, and Jesperson.

However, of all the descriptivists, special mention should be givento Otto Jespersen (1860 – 1943), a Danish linguist whose most enduringwork is in the theory of grammar and the grammar of English. Like Sweet,he proposes three principles of classification – meaning, form, and func-tion. He is much more original in syntactic studies. His theory is set outin The Philosophy of Grammar (1924). It is based on the concepts ofranks distinguished in nexus (predication) and junction (subordination).

The term rank is used of successive levels of subordination, or depen-dency. E.g. in the junction very cold water, water has the highest rank and isa primary; cold has the next highest rank and is a secondary; very has thelowest and is a tertiary. The ranks are also distinguished in nexus, e.g. He(primary) writes (secondary) a letter (primary) every day (tertiary). This sen-tence contradicts his theory of ranks since a letter is subordinate to writes. Ifthe scholar had been more consistent, he would have had to apply the sameprinciple of subordination to both junction and nexus structure, as he did inhis analysis of a furiously barking dog and a dog barks furiously.

Despite this inconsistency, the theory of ranks undoubtedly servedas an impulse to transformational-generative grammarians who saw trans-formational relations between predicative and non-predicative structures.Non-predicative structures were treated as transformationally derived fromthe corresponding predicative ones – both were built on the same type ofsubordination.

Summary: 1. Unlike prescriptivists, descriptivists focus their atten-tion on actual usage without trying "to settle the relative correctness ofdivergent usages". 2. Descriptivists rely on the English of the best au-thors of their day as well as the English of the past. To them, change inlanguage is not associated with corruption; 3. Similar to prescriptivists,descriptivists use meaning and function in their definition of parts ofspeech.

Check yourself test 21. Why did scholars make repeated attempts to group languages

according to their origin?How many languages are there in the world? Is the number of them

growing or diminishing? How do they differ from one another?What scholars were the first to develop the idea of language rela-

tionship?

31

2. In connection with what studies was the historical comparativelinguistics created?

What stimulated the appearance of the historical comparative meth-od in linguistics?

Who was the first to put forward a scientific hypothesis that San-skrit was related to some European languages?

How did Sir William Jones ground his ideas?What scientists are the creators of the historical comparative method?What languages did they study?What was the principal line of European linguistics in the 19th century?Who continued to develop the historical comparative method in the

19th and 20th centuries?3. What are the aims of the historical comparative method?What lingual elements are called 'prehistoric'?What is meant by the term 'reconstruction'?For what other purposes is the historical comparative method used

besides reconstructing prehistorical forms?4. How do families of related languages arise?What causes the splitting of languages?What epoch in the history of peoples is characterised by language

divisions?What language is called the 'parent language'?Can the parent language continue to exist in living speech after a

family of languages has arisen out of it?Do all the kindred languages survive after the splitting of the parent

language?Why is language not a racial concept?5. What features of resemblance between languages prove their kinship?Why is the arbitrary character of the lingual sign so important for

the historical comparative method?What resemblances between languages do not evidence their kinship?Why must we compare the elements of the basic word stock and

grammatical affixes to prove that the languages are kindred?Give examples of related native words in the Indo-European lan-

guages.What is the system of phonetical correspondences between kindred

languages?What do the phonetical correspondences show?6. Why is it important for the comparative study of languages to

note the uneven character of language development?Why do languages develop unevenly? How are reconstructions

formed?

32

What notation is used to represent historical changes in scientifictreatises? Give examples.

7. What are the main limitations of the historical comparative method?What material is analisable by the historical comparative method?Are all the common features of related languages inherited directly

from the parent language?What is dangerous in elaborating reconstructions?Can the historical comparative method be applied to spoken lan-

guages?Is the historical comparative method antiquated?What other families of languages besides the Indo-European are

being studied by the comparative linguistics?

1.6. THE FUNDAMENTAL IDEAS AND MAIN SCHOOLSOF MODERN LINGUISTICS

The main method of the 19th century and the beginning of the 20th

century was the historical comparative method. Valuable as it was for thescientific study of languages, it had definite shortcomings and limitations.

The historical comparative method gave no exact definition of theobject of linguistics as an independent science. Logical, psychological,and sociological considerations were involved in linguistic studies to suchan extent as to obscure linguistics proper. As Louis Hjelmslev rightlypointed out, "The linguistics of the past – even of the recent past – hasconcerned itself with the physical and physiological, psychological andlogical, sociological and historical precipitations of languages, not of thelanguage itself.

The study of numerous languages of the world was neglected, theresearch being limited to the group of the Indo-European languages. Theresult stated by Hjelmslev in 1939 was that a single morphological type –a type unique in the world – determines the structure of the theory.

It was mainly the historical changes of phonological and morpho-logical units that were studied; syntax hardly existed as an elaborate do-main of linguistics alongside of phonology and morphology. The pains-taking study of the evolution of sounds and morphemes led to an atomis-tic approach to language.

As a reaction to the atomistic approach to language a new theory ap-peared that was seeking to grasp linguistic events in their mutual interconnec-tion and interdependence, to understand and to describe language as a system.

33

The first linguists to speak of language as a system or a structure ofsmaller systems were Beaudouin de Courtenay (1845 – 1929) and Acad.F.F.Fortunatov (1848 – 1914) of Russia and the Swiss linguist Ferdinandde-Saussure (1857 – 1913).

The work that came to be most widely known is de-Saussure's "Coursde linguistique generale" (Course in General Linguistics), posthumous-ly compiled from his pupils' lecture-notes between 1906 and 1911.

De-Saussure's main ideas, which are acceptable to us with certainreservations and explanatory remarks, are as follows:

1. Language is understood as a system of signals (linguistic signs),interconnected and interdependent. It is this network of interdependentelements that form the object of linguistics as an independent science;

2. Language as a system of signals may be compared to other systems ofsignals, such as writing, alphabets for the deaf-and-dumb, military signals,symbolic rites, forms of courtesy, etc. Thus, language may be considered asbeing the object of a more general science – semiology – a science of the futurewhich would study different systems of signals used in human societies.

3. Language has two aspects: the system of language (French:langue) and the manifestation of this system in social intercourse – speech(French: parole). The system of language is a body of linguistic units –sounds, affixes, words, grammar rules and rules of lexical series. Thesystem of language enables us to speak and to be understood since it isknown to all the members of a speech community. Speech is the total ofour utterances and texts. It is based on the system of language, and itgives the linguist the possibility of studying the system. Speech is thelinear (syntagmatic) aspect of languages, the system of language is itsparadigmatic ('associative' as de-Saussure called it) aspect.

4. The linguistic sign is bilateral, i.e. it has both form and meaning.We understand the meaning of the linguistic sign as reflecting the ele-ments (objects, events, situations) of the outside world.

5. The linguistic sign is 'absolutely arbitrary' and 'relatively moti-vated'. This means that if we take a word 'absolutely' disregarding itsconnections to other words in the system, we shall find nothing obligato-ry in the relation of its phonological form to the object it denotes (ac-cording to the nature of the object). This fact becomes evident when wecompare the names of the same objects in different languages, e.g.:

English: ox hand winterFrench: boeuf main hiverRussian: бык pyкa зимaThe 'relative motivation' means that the linguistic sign taken in the

system of language reveals connections with other linguistic signs of thesystem both in form and meaning. These connections are different in dif-

34

ferent languages and show the difference of "the segmentation of the pic-ture of the worl"' – difference in the division of one and the same objectivereality into parts reflected in the minds of different peoples, e.g.

English: arrow – shoot; apple – apple-treeRussian стрела – стрелять; яблоко – яблоня6.Language is to be studied as a system in the 'synchronic plane',

i.e. at a given moment of its existence, in the plane of simultaneouscoexistence of elements. We understand the synchronic plane as a givenmoment (of more or less longer duration) of the historical developmentof the language studied.

7. The system of language is to be studied on the basis of theoppositions of its concrete units. The linguistic elements (units) can befound by means of segmenting the flow for speech and comparing theisolated segments, e.g. in "the strength of the wind" and in "to collectone's strength" we recognise one and the same unit 'strength' in accordwith its meaning and the form; but in 'on the strength of this decision' themeaning is not the same and we recognise a different linguistic unit.

There were three main linguistic schools that developed these newnotions concerning language and linguistics as the science that studies it:the Prague School that created Functional linguistics, the CopenhagenSchool which created Glossematics, and the American School that createdDescriptive linguistics. The Immediate Constituents Grammar was afurther development of descriptive linguistics; the TransformationalGrammar, the latest.

The Prague SchoolThe Prague School was founded in 1929 uniting Czech and Russian

linguists: Mathesius, Trnka, Nikolay Trubetzkoy, Roman Jakobson and others.The main contribution of early Praguians to modern linguistics is

the technique for determining the units of the phonological structure oflanguages. The basic method is the use of oppositions (contrasts) ofspeech-sounds that change the meaning of the words in which they occur.The basic definitions are given by Trubetzkoy as follows:

Rule 1: If in a language two sounds occur in the same position andcan be substituted for each other without changing the meaning of theword, such sounds are optional variants of one and the same phoneme.

Rule 2: If two sounds occur in the same position and cannot besubstituted for each other without changing the meaning of the word ordistorting it beyond recognition, these two sounds are phonetic realisationsof two different phonemes.

Rule 3: If two similar sounds never occur in the same position, theyare positional variants of the same phoneme.

35

Trubetzkoy developed an elaborate set of contrast criteria for theidentification (recognition) and classification of phonological oppositions.The most widely known is the binary 'privative' opposition in which onemember of the contrastive pair is characterised by the presence of a certainfeature which is lacking in the other member (hence 'privative'). Theelement possessing the feature in question is called the 'marked' (strong)member of the opposition, the other is called the 'unmarked' (weak)member of the opposition. A phoneme is distinguished from all the otherphonemes by a set of distinctive (differential) features, e.g. [p] isdistinguished from [b] as a voiceless sound, from [t] as a bilabial, from[m] as having no nazalisation, etc. Thus any phoneme is defined as a setor 'bundle' of differential (distinctive) features.

Trubetzkoy stressed the fact that his technique of analysis may beused in other domains of linguistics. The method of oppositions has beensuccessfully extended to grammar and semantics.

Oppositional analysis in morphologyThe principle of binary oppositions is especially suitable for

describing morphological categories.Binary relations penetrate practically every plane of language –

phonological, morphological, and syntactic, but are especially evidenton the morphological level, which better than any other reflects thestructural organisation of a particular language, its intricate correlationsand the interdependence of its units.

The principle of privative oppositions has been used by Roman Jakobsonfor describing the morphological categories of the Russian language.

One of the most interesting examples is the description of the Russiancase system in terms of binary privative oppositions. Jakobson proposedthe following three distinctive features: A – direction, B – objectiveness,C – periphery. The result is represented in the following table:

Distinctive features

Cases

А

В

С

Nominative Genetive Dative Accusative Instrumental Prepositional

- - + + - -

- + - - - +

- - + - + +

36

where + (plus) means the presence of the feature in question, thuscharacterising the corresponding case as the marked member of theopposition, and – (minus) indicates the absence of the feature in question,thus characterising the corresponding case as the unmarked member ofthe opposition. Thus the three rather abstract distinctive features proposedby Jakobson form 'bundles' one and only one of which is typical for eachof the 6 Russian cases.

The principle of privative oppositions can be easily applied to Englishmorphology.

The most general case is that of the general system of tense-forms ofthe English verb. According to Prof. N. F. Irtenyeva's theory, the tense-forms of the English verb are divided into two halves: that of the tense-forms of the present plane, and that of the tenses of the past plane. Theformer comprises the Present, Present Perfect, Present Continuous, PresentPerfect Continuous, and the Future tense-forms; the latter includes Past,Past Perfect, Past Continuous, Past Perfect Continuous and the Future-in-the-Past. The second half is characterised by specific formal features –either the suffix -ed (or its equivalents) appear, or a phonemic modificationof the root. The past is thus a marked member of the opposition 'present–past' as against the present sub-system, which is the unmarked member.

Prof. B. Ilyish points out that "the opposition between perfect andnon-perfect forms is shown to be that between a marked and an unmarkeditem, the perfect forms being marked both in meaning (denotingprecedence) and in morphological characteristics 'have + secondparticiple', and the non-perfect forms – unmarked both in meaning(precedence not implied) and in morphological characteristics (purelynegative characteristic: the collocation 'have + second participle' notused)" [12, p. 132].

The obvious opposition within the category of voice, as Prof. Ilyishrightly points out, is that between active and passive. A few pairs of parallelforms involving different categories of aspect, tense, correlation, and moodillustrate the opposition of active/passive, namely:

invites is invited is beingis inviting invited was invited hasinvited has been invited should be

invited"From the point of view of form," Prof. Ilyish remarks, "the passive

voice is the marked member of the opposition: its characteristic is thepattern 'be + second participle', whereas the active voice is unmarked: itscharacteristic is the absence of that pattern" [12, p. 133].

37

Oppositional analysis in syntaxThe principle of privative oppositions has been recently used to

represent the traditional sentence-parts of the basic (independent declarativenon-emotional) two-member sentence type. The parts of such a sentencetype are defined by their position in the structure of the sentence: the subjectto the left of the verb-predicate, the object to the right of the verb, theadverbial modifier to the right of the object (if any); the attribute, that mayappear as an optional sentence-part, occupies the position in contact to thenoun. The syntactic relations of the sentence parts are characterised bythree distinctive features: A – subordination, B – predicativeness(appurtenance to the group of the verb-predicate), and C – objectiveness –feature connected, but not without reservation, with the possibility ofchanging the active to the passive construction.

Thus we have:

An application of the oppositional method has also been extendedto describe different types of simple sentences in Modern English. Therules are as follows:

Rule 1. Different sentence-types are those that cannot be substitutedfor each other without changing the structural meaning of the sentence.Here belong:

(a)two-member sentences as against one-member sentences, e.g."John worked" as against "John!" or "Work!";

(b)sentences differing in the arrangement of the main constituents inbasic sentences, e.g. "We saw a river there" as against "There is a river there";

(c)sentences differing in the case-form of the subject-noun, e.g.:"Mary was a happy girl" as against "Mary's was a happy life".

Rule 2. Variants of one and the same sentence-type are those thatcan be substituted for each other without changing the structural meaningof the sentence or distorting it beyond recognition. The following variantsare recognised:

Distinctive features \ sentence parts A B C

Subject . . . – – –

Predicate . . . – + –

Attribute . . . + – –

Object . . . + + + Adverbial modifer . . .

+ + –

38

(a) positional variants – context sensitive sentences in which one ormore elements are left out but can be unambiguously inferred from thepreceding sentence. There are two kinds of positional variants:

Included positional variants – such as can be placed in the positionoccupied in the. preceding sentence by a question word or a word whichis repeated in the positional variant, e.g.: "Who gave you that? Soames.""Where did she see him? – In the park." "What do you think I am madeof? Leather?" "Soames gave it her."Who?" etc.

Adjoined positional variants – such as can be optionally added tothe preceding sentence, e.g. "I am leaving. Tonight. Immediately."

(b)Optional variants – extended sentences as against unextendedsentences, the unextended sentences being understood as having objects,etc., in accord with the valence of the verb (necessary to make the sentencecomplete), e.g.: "She saw him" and "She saw him yesterday in the park";"Put these things on the table" and "Put these things on the tableimmediately", etc.

(c)Stylistic variants, which may be:emotional: "I saw him!" "She is such a darling!"colloquial: "You done it." "You going to work here again?""Father in town?" "Lost my job, Vic." "Ever had practice?", etc.

The Copenhagen SchoolThe Copenhagen School was founded in 1933 by Louis Hjelmslev

(1899 – 1959) and Viggo Brandal (1887 – 1942). In 1939 the Prague andthe Copenhagen Schools founded the magazine Acta Linguistica that hadbeen for several years the international magazine of Structural Linguistics.In the early thirties the conception of the Copenhagen School was giventhe name of "Glossematics" (from Gk. Glossa – language).

In 1943 Hjelmslev published his main work "Omkring SprogteoriensGrundlaeggelse" (Principles of Linguistics) which was translated intoEnglish and appeared in Baltimore in 1953 under the title Prolegomenato a Theory of Language. A Russian translation was published in 1960 inthe first volume of «Hoвoe в лингвистике».

Glossematics sought to give a more exact definition of the object oflinguistics. The two sides of the linguistic sign recognised by de-Saussureare considered by Hjelmslev to have both form and substance. This leadsto the recognition of a bilateral character of the two planes – 'the plane ofcontent' and 'the plane of expression'.

The object of the linguistic science is limited to the two inner layers –the form in the plane of content and the form in the plane of expression,i.e., linguistics studies nothing but form.

39

The form in the plane of content is the segmentation (the division)of the picture of the world, which is different in different languages, e.g.

English: blue foot, leg hand, armRussian: синий, голубой нога рукаSimilar differences may be easily found in tense and case-systems,

in the expression of genders in different languages, etc.The two inner layers are connected by the 'law of commutation,'

which means that differences in the plane of expression signal differencesin the plane of content. There is no simple one-to-one correspondence ofthe two planes. The units of the planes may be decomposed into smallercomponents which reveal the correspondences of the two planes.Hjelmslev sought to frame a sort of linguistic calculus (исчисление)which might serve linguistics in the same way as mathematics servedphysical sciences. The object of linguistics was then understood as'language in the abstract'– not as concrete natural languages. Hjelmslev'sideas which are now extensively used in the study of semantics, are atthe basis of the componential analysis.

Check yourself test 3Part 1

1. What was the main method of linguistics in the 19th and thebeginning of the 20th centuries?

Was the object of linguistics rigorously defined at that time?What languages were mainly studied by the historical comparative

method?What linguistic units were studied? Did syntax exist as a domain of

linguistics?Were the linguistic elements studied as a system? What is the

'atomistic approach' to language?2. What linguists were the first to speak of language as a system?What work came to be widely known in this connection?Can we accept de-Saussure's ideas without reservation?How is language understood by de-Saussure?What is the object of linguistics as an independent science according

to de-Saussure's theory?3. What does 'bilateral' mean in reference to the linguistic sign?How do we understand the meaning of the linguistic sign?What is the 'synchronic plane'?What are the two aspects of language?How did 'de-Saussure understand the associative (paradigmatic)

series in the system of language?

40

In what sense is the linguistic sign 'absolutely arbitrary'?What is the 'relative motivation' of the linguistic sign?Give examples to show that the linguistic sign is absolutely arbitrary.Give examples to show that the linguistic sign is relatively motivated.What is semeiology and what semeiotic systems do you know?4. How can we study the system of language? How can we segment

the flow of speech?What must we do besides segmenting the flow of speech to recogn-

ise the units of a language?5. What are the main schools in modern linguistics?6. By whom was the Prague School founded and when?What is the contribution of early Praguians to modern linguistics?What is the basic method of research used by Trubetzkoy?What are the three rules given by Trubetzkoy?What is 'privative opposition'? What is the 'marked' and the 'un-

marked' member of a privative opposition?Can Trubetzkoy's technique of analysis be used in other domains of

linguistics?7. What is the role of binary oppositions in the description of mor-

phological categories?How did Roman Jakobson describe the Russian case-system?How can the idea of privative binary oppositions be applied to the

general English tense-system?What is the opposition between perfect and non-perfect forms of

the English verb?What is the opposition found in the system of the English voice?8. How do we distinguish different sentence-types?What are the main criteria for distinguishing different sentence-types?What is the general characteristic of the variants of one and the

same sentence-type?What are positional variants of a sentence?What are the two types of positional variants?What are the optional variants of a sentence-type?What are the stylistic variants of a sentence-type?Give examples of emotional and colloquial stylistic variants.What is the criterion for distinguishing a variant from a different

sentence-type?9. By whom was the Copenhagen School founded?What is Hjelmslev's main work?What was the role of glossematics in defining the object of lin-

guistics?

41

How does Hjelmslev understand the structure of language?What is the object of linguistics according to Hjelmslev?What does 'form' mean in the plane of content?What does 'form' mean in the plane of expression?What is the 'law of commutation'?Is 'linguistic calculus' meant for concrete languages?

Part 21. Give Russian names of the following things:table, chair, book, pencil, horse, dog, dress, hat, boy, girl and give

other examples to show that the linguistic sign is absolutely arbitrary.Think of any other languages you know.

2. Translate into Russian:horse, horse and foot, horse-path, top, the top of the class, at the top

of one's voice, room at the top hand, sleeve, tame word, closing speech,not a syllable, according to his story, in short

3. Give other examples to show the relative motivation of linguisticsigns in different languages.

Find the Russian equivalents for the English: time, tense; table;clock, watch; meal; 24 hours and the English equivalents for the Rus-sian: человек, дерево, сутки, дело.

4. Compare the English Conditional Mood with the Russian Sub-junctive (cocAaeameAbHoe naKAOuenue) and state the difference ingrammatical meanings.

5. Prove that the English Possessive Case is not equivalent to theRus-sian Genitive (poдительный падеж).

6. Translate into Russian:1. He cut his finger. You finger this and find that. 2. He walked

round the house. He failed to round the lamp-post. He took his dailyround. The ladder has many rounds.

7. Give other examples to show the different segmentation of "thepicture of the world" in different languages.

8. Find the positional variants of simple sentences in the followingexamples: 1. "Tom!" – No answer. 2. She shuddered. "Horrible weather,"she сommented. 3. "Whatever induced him to do such a dreadful thing?"–"The climate." 4. "Water! For Heaven's sake, water" 5. A knock at thedoor. "Your hot water." 6."Do you want roast beef or tongue?"– "Roastbeef." 7. "Who told you that? Harold?"

42

9. Find the colloquial variants:1. "It's getting dark," she said. "Be dark in half an hour," Harry said.

2. "I've never been there, you know." – "Been in India?" 3. "Being noblenow, Olwen? You needn't, you know."

10. Find the included and the adjoined positional variants:1. "So here I am. For a few days," he added. 2. "What do you do

when you are on leave? Play golf? Sail a boat? Go fishing?" 3."Where doyou come from? –"Paris." 4. "I may get married."To Celeste?" 5. Whatam I? Her uncle? 6."Where is he?" – Here. 7. Harold was always veryabstentious." – "Here," said the widow. Did you drink? – Like a fish.8. Did you come to see your aunt? – No, to see my uncle." 9. "Do you likeBart, dear?" – Like him? We have nothing in common." 10. "This so-called town of yours hasn't any width. Just length.".

1.7. STRUCTURAL DESCRIPTIVE GRAMMAR

The non-structural descriptive approach to language had its heydaybetween 1900 and 1930, when it was replaced by structuralism. The fa-ther of American structuralism is generally called Leonard Bloomfield,who in his book Language presented the new approach as follows: "Thestudy of language can be conducted... only so long as we pay no attentionto the meaning of what is spoken" [2].

Descriptive linguistics developed from the necessity of studying half-known and unknown languages of the Indian tribes.

At the beginning of the 20th century these languages were rapidlydying out under the conditions of what is known as 'American culture', or'American way of life', which had brought the Indian peoples poverty,diseases and degradation. The study of these languages was undertakenfrom purely scientific interests.

The Indian languages had no writing and, therefore, had no history.The comparative historical method was of little use here, and the first stepof work was to be keen observation and registration of linguistic forms.

Furthermore, the American languages belong to a type that has littlein common with the Indo-European languages; they are 'agglomerating'languages, languages devoid of morphological forms of separate wordsand of corresponding grammatical meanings. Descriptive linguists hadtherefore to give up analysing sentences in terms of traditional parts ofspeech; it was by far more convenient to describe linguistic forms ac-cording to their position and their co-occurrence in sentences.

The description of a language became more refined at the beginningof the 20th century due to the development of the concept of 'phoneme'. The

43

concept of phoneme was worked out by the Russian linguists Baudouin deCourtenay and his student Kruszewski, and developed by the linguists ofthe Prague School (Roman Jakobson, 1928; Trubetzkoy, 1939).

Franz Boas, linguist and anthropologist (1858 – 1942) is usuallymentioned as the predecessor of American Descriptive Linguistics. Hisbasic ideas were later developed by Edward Sapir (1884 – 1939) andLeonard Bloomfield (1887 – 1949). Bloomfield's main work Languagewas published in 1933. All linguists of the USA at one time or other feltthe influence of this book.

The American linguists began by criticising the Praguian opposi-tional method and claiming a more objective – distributional – approachto phonemes. But it soon became clear that the facts established were thesame, and only the approach was different. The American DescriptiveSchool (if we may speak of a number of loosely interrelated approachesas a 'school') began with the works of Edward Sapir and Leonard Bloom-field. American linguistics developed under the influence of these twoprominent scientists.

Sapir studied a great variety of languages (Indian and Malayo-Polynesian). His most known work is Language. An Introductory to theStudy of Speech (1921). Leonard Bloomfield is considered to be a morerigid theorist. His book of the same title as Sapir's Language, is moresystematic than Sapir's, and the treatment of linguistic problems is moremodern. It is a complete methodology of language study, approachingthe language as if it were unknown to the linguist (student). The ideaslaid down in this book were later developed by Z.S.Harris, Ch. C.Friesand other contemporary linguistic students.

Followers of this approach sought to study the structure of a lan-guage as objectively as possible, without reference to meaning and otherlanguages. By other languages they, first and foremost, meant Latin andGreek, the languages prescriptive and, to a lesser degree, descriptive gram-marians modelled their analysis on. English was regarded as a languagehaving its specific structure, and the task of a linguist was to reveal it byusing scientific (i.e. formal) methods of analysis.

Meaning as a criterion was not reliable since, being unobservable,it could be interpreted differently by different linguists. Therefore thelinguist was to devise formal methods of analysis and replace meaningby form; the linguist must be interested in what he observes, i.e. objec-tive data. The structuralists based their conclusions on the analysis ofsentences that they had collected from native speakers of English, givingpriority to Spoken English.

To structuralists, language is a highly organized affair, where thesmaller units are built into larger units, which in turn are built into larger

44

ones, until the largest unit is reached. Such building-blocks are phonemesand morphemes. The structures that we build out of the ‘bricks’ are lex-emes. Lexemes, in their own turn, serve to build the largest unit, thesentence, i.e. the predicative unit.

Structural linguists ignored meaning not because they were not in-terested in it. Meaning was ignored on the grounds that it was not ob-servable and could not be described objectively by using formal meth-ods. The description of meaning had to wait until appropriate methodswere devised. Such being the case, they focused their attention on struc-tural, i.e. grammatical, meaning.

How are structural meanings conveyed in English? Structural gram-marians have pointed out four devices used in English to indicate struc-tural meaning: 1.word form; 2) function words; 3) word order; 4) intona-tion and accent patterns (prosodic patterns).

To have a deeper understanding of modern grammar we must getacquainted with the main concepts of Bloomfield's book.

1. Bloomfield understood language as a workable system of signals,that is linguistic forms by means of which people Communicate: "...everylanguage consists of a number of signals, linguistic forms" [2, p. 158].

2. Bloomfield's understanding of 'meaning' seemed to be very un-usual at that time. Later his concept of 'meaning' was developed by CharlesFries but even now 'meaning' is one of the problems linguistics seeks tosolve. Bloomfield says: "...by uttering a linguistic form a speaker promptsthe hearers to respond to a situation; this situation and the responses to it,are the linguistic meaning of the form. We assume that each linguisticform has a constant and definite meaning, different from the meaning ofany other linguistic form in the same language." [2, p. 158]. Accordingto Bloomfield, "the meanings of speech-forms could be scientifi-callydefined only if all branches of science, including, especially, psychologyand physiology, were close to perfection. Until that time, phonology, andwith it all the semantic phase of language study, rests upon an assump-tion, the fundamental assumption of linguistics: we must assume that inevery speech community some utterances are alike in form and mean-ing." [2, p. 78]. And then: "...every utterance contains some significantfeatures that are not accounted for by the lexicon." (p. 162). "...No matterhow simple a form we utter and how we utter Mr... the utterance conveysa grammatical meaning in addition to the lexical content..." [2, p. 168].The quotations clarify two things: (1) the meaning of an utterance can befound through the response of the hearers; (2) a sentence has a grammat-ical meaning which does not (entirely) depend on the choice (selection)of the items of lexicon.

45

This can be illustrated by the following:1. The selection of 'none' instead of 'someone' changes the meaning

of an affirmative statement into negative ("Someone has come – Nonehas come"); the selection of an animate noun instead of the inanimate ispossible only with a changed meaning of the verb:

The wind blew (the leaves away) – The man blew his nose."Bloomfield's idea that the meaning of a sentence is part of the mor-

pheme arrangement, and does not entirely depend on the words used inthe sentence has later been developed by Ch.Fries and N.Chomsky, whoshowed that sentences with non-sensical selection of words still have adefinite meaning because of the arrangement of linguistic forms.

2. Bloomfield understood grammar as meaningful arrangement oflinguistic forms from morphemes to sentences. He wrote that the mean-ingful arrangement of forms in a language constitutes its grammar, andin general, there seem to be four ways of arranging linguistic forms:(1) order; (2) modulation: "John!" (call), "John?" (question), "John" (state-ment); (3) phonetic modification (do – don't); (4) selection of forms whichcontributes the factor of meaning.

3. He produced the definition of the sentence that is now acceptedby modern American linguistics. This definition is given in Ch.Fries'sbook The Structure of English as the best among other two hundred def-initions, and it reads as follows: "...Each sentence is an independent lin-guistic form, not included by virtue of any grammatical construction intoany larger linguistic form" [2, p. 136].

This definition is essentially like that of another great linguist ofthat period, that is Otto Jespersen, the Great Dane, as he is sometimesreferred to. Bloomfield also said that a sentence is marked off by a; cer-tain 'modulation' or intonation. He stated that in English the most favou-rite type of sentence is the 'actor – action' construction having two posi-tions. These positions are not interchangeable. All the forms that can fillin a given position thereby constitute a form-class. In this manner thetwo main form-classes are detected: the class of nominal expressionsand the class of finite verb expressions.

4. Thus Bloomfield has shown a new approach to the breaking up ofthe word-stock into classes of words, the positional approach. Bloomfieldwrites: "The syntactic constructions of a language mark off large classes offree forms, such as, in English, the nominative expression or the finite verbexpression..." "We shall see that the great form-classes of a language aremost easily described in terms of word-classes (such as the traditional partsof speech), because the form-class of a phrase is usually determined byone or more of the words which appear in it." [2, p. 190].

46

These long form-classes are subdivided into smaller ones. In mod-ern linguistic works the nominal phrase of a sentence is marked as thesymbol NP, and the finite verb phrase – VP. The symbols N and V standfor the traditional parts of speech, nouns and verbs, although the NP mayinclude not only nouns but their equivalents and the noun determiners(e.g.: the man, my hand, this house, I, they, something, some, others,etc.); and the VP with a transitive verb may have a NP in it (took a book,sent a letter, etc.). The long form-class of N is now subdivided into: ani-mate and inanimate, material and abstract, class nouns and proper nouns.The long form-class of V is subdivided into intransitive verbs (VI), tran-sitive verbs (VT) and the latter are again divided into the V of the 'taketype', the 'give type', the 'put type’ and the 'have type', etc.

The division of the word-stock into form-classes is developed inFries' book The Structure of English and is dealt with in a most knownarticle by Zellig S.Harris Co-Occurrence and Transformation in LinguisticStructure.

5. Perhaps Bloomfield was one of the first to speak about 'utter-ance' as a linguistic unit. Meanwhile the concept of utterance is of impor-tance for the study of unknown languages, and this concept also eluci-dates many syntactic problems in the familiar languages. The concept ofutterance is now accepted by the linguists of different schools.

6. The first mentioning of the Immediate Constituents (IC) can alsobe found in Bloomfield's book. This theory of the IC which in the middleof the 20th century fascinated the minds of the linguists, and has onlybeen obscured by the Transformational grammar, was first propoundedby Bloomfield. We may suppose that the idea of the IC arose under theinfluence of Panmi's grammar because in the first chapter of his bookBloomfield says that Panini's grammar taught the Europeans to study theIC of their languages

These are the main ideas of Bloomfield's Language which make thebook a predecessor of American Descriptive linguistics.

Present-day English depends strongly on word order to convey mean-ing. Charles Fries argues that "certain positions in the English sentencehave become to be felt as subject territory, others as object territory, andthe forms of the words in each territory are pressed to adjust themselvesto the character of that territory" [8, p. 158]. Function words are anotherdevice. Having little or no lexical meaning of their own, they serve tovary the functions of the lexical words. Consider: The mother of the boywill arrive tomorrow. The words mother, boy, arrive, and tomorrow havemeaning in themselves quite apart from their grammatical relation, ormeaning, in the sentence. They are full, or notional, words. But the wordsthe, of, and will express primarily a grammatical idea and have little or

47

no meaning apart from the grammatical function they indicate: the func-tions as a determiner of mother telling us that a particular member of theclass is meant; of relates the boy to the mother or, in other words, ofmakes the word boy an attribute, or modifier, of the word mother; it isequivalent to a genitive inflexion (cf. the boy’s mother); will indicatesthat the process of arriving will occur in the future.

The role of intonation is obvious when we have to differentiate be-tween statements and questions, between the theme and the rheme. Stress,or accent, helps to distinguish nouns from verbs (e.g. .suspect vs. sus-pect), juncture-pause in speech distinguishes between such structures asnight-rate and nitrate or phrases, clauses and sentences.

As already mentioned, anxious to be objective, structural grammari-ans used formal methods of linguistic analysis, such as immediate constit-uent, distribution, substitution, transformation (deletion, permutation, etc.).

The term immediate constituents (IC) was introduced by L.Bloomfieldas follows: "Any English-speaking person who concerns himself with thismatter is sure to tell us that the immediate constituents of Poor John ranaway are the two forms Poor John and ran away; that each of these is, inturn, a complex form; that the immediate constituents of ran away are ranand away, and that the constituents of Poor John are poor and John".

To put it in more simple language, the constituents Poor John andran away belong together, for they stand side by side. They are the mostimportant constituents since they constitute the core of the sentence. Thesame principle of togetherness underlies the constituents Poor and John,ran and away. However, as compared to Poor John and ran away, theyare constituents of a lower level: they are subconstituents of the higherlevel – Poor John and ran away. Hence two levels of analysis: higherand lower where the lower level is subordinated to the higher level.

According to D. Bolinger, the principle of togetherness is very perva-sive in language. It manifests itself in "our resistance to putting somethingbetween two things that are more closely related to each other than theyare to what is inserted. Teachers find it hard to enforce the rule of interiorplurals in forms like mothers-in-law and postmasters general – speakerswant to put the –s at the end. They are even more reluctant to say hardest-working person, inserting the –est between the members of the compoundhard-working; and though some might manage it there, probably no onewould say farthestfetched story for most far-fetched story" [12, p. 158].

The aim of IC analysis is to discover and demonstrate the interrela-tionships of the words in a linguistic structure – the sentence or the word-combination. It is not difficult to see a similarity between immediateconstituent analysis and the traditional procedure of ‘parsing’ sentencesinto subject and predicate, attribute, object and adverbial.

48

Thus L.Bloomfield’s sentence could be described by a traditionalgrammarian as a simple sentence whose subject is a noun phrase, madeup of the noun John modified by the adjective poor, and whose predicateis a verb-phrase consisting of the verb ran modified by the adverb away.

Both the traditional procedure and the IC method view the sentencenot as just a linear sequence of elements but as made up of "layers" ofimmediate constituents, each lower-level constituent being part of a high-er-level constituent.

As already mentioned, the aim of IC analysis is to show the syntag-matic interrelations between the sentence constituents. Structuralistswould agree that if we have described these interrelationships, we havedescribed the syntax of the sentence in its entirety.

The shortcoming of the IC method lies in its extreme formality: theanalyst, using this method, is not interested in the content of the interrela-tionships. Such syntactic notions as subject, predicate, object, complement,attribute, adverbial, which constitute the basis of traditional analysis, prac-tically were never used by structuralists. In this way, content was separatedfrom form. And language is a dialectical unity of content and form.

Besides, the method of IC analysis is only capable of revealing wordrelationships within the sentence. But "How could a frame so confinedas that of immediate constituents be expected to fit comfortably aroundthe whole of syntax, when there are many important relationships thatescape it? The classic example is the relationship between the active andthe passive voice: George sees Mary; Mary is seen by George. An imme-diate-constituent analysis of these two sentences tells nothing about theirunderlying kinship." [12, p. 158].

Let us now turn to distribution.Distribution is the set of contexts, or environments, within sentenc-

es in which a unit can appear. So, for instance, the distribution of hair inwritten English is the set of the following contexts:

I combed my hair.Give me the hair spray.My hair is too long, etc.The distribution of the word hair can be described as follows: 1) it

can follow the word my; 2) it can precede the word spray; 3) it can pre-cede the verb be.

If we analyze other words, we shall find other positions they occu-py, or other environments in which they are used. Words that have thesame distribution are words of the same class. We test their distributionby substituting them for other words. Consider the sentence I combed myhair. The word hair can be formally substituted for other words, such asplace, town, wood, etc.

49

Distribution and substitution were used by structuralists for the clas-sification of linguistic units. Like the IC method, the method of distribu-tion was treated as a method that enables the analyst to classify wordsinto classes objectively, i.e. without having recourse to meaning.

The transformational method was developed by Zellig Harris in the 1950s.The aim of a transformational operation was to reveal similarities and differ-ences in the structure of the units being examined or to reveal the structuralpotential of the unit. To understand it, let us examine the following structures:

1) Mary has a new car.2) Mary has a good time.Superficially, the two sentences are identical in structure. However,

they present two distinct structures. Sentence (1) cannot be turned intothe passive while sentence (2) can:

Mary has a new car. – A new car is had by Mary.Mary has a good time.– A good time is had by Mary.The structural potential of a linguistic unit can also be tested by this

method:a) my dog – the dog of mine;b) Susan’s dog – the dog of Susan – the Susan dog;c) John gave the book to me. – John gave me the book. – The book

was given to me.d) John bought the book for me. – John bought me the book. – The

book was bought for me. – I was bought the book.e) A number of people came. – People came – A number came. –

The number of people came.f) Bill fixed up a drink for John. – Bill fixed a drink up for John. –

Bill fixed a drink for John up. – Bill fixed up John a drink.No matter which aim is pursued, transformations help to reveal the

existing relations between linguistic structures.g) John resides in New York. – John resides.h) John is my best friend. – John is.i) John is walking in the park. – John is walking.j) Mary put the flowers in the vase. – Mary put the flowers.k) Mary is writing a letter. – Mary is writing.1) The door was closed. – the door was closed by the janitor.m) The door closed. – The door closed by the janitor.n) The woman looked angry. – The woman looked angrily.o) The woman appeared angry. – The woman appeared angrily.p) We do not allow smoking in the lecture hall. – It is not allowed to

smoke in the lecture hall. – Smoking is not allowed in the lecture hall.

50

q) The student arrived late. – The student’s late arrival.Through the transformational method we can show the structural

potential of a linguistic unit as compared to units exhibiting superficiallysimilar structure. If linguistic units can be subjected to the same transfor-mation, we can say that they are identically structured. But if they can-not, their structure is different.

Тo sum up, the merit of the transformational method can be statedas follows: 1) it enables the analyst to diagnose linguistic structures; 2) itreveals the structural potential of linguistic structures. The emergence ofthis method practically marks the end of post-Bloomfieldian linguisticsand the beginning of a new stage of structural linguistics.1

Check yourself test 41. What gave rise to the advent of Descriptive linguistics?Why was the comparative historic method of little use in the study

of the languages of Indian tribes?What was the first step of work in this field?What type do American Indian languages belong to?What does the term 'agglomerating language' mean?What was the convenient method for describing such languages?2. What concept contributed to a more refined description of lan-

guages?By whom was the concept of phoneme worked out?Whom do we usually mention as the predecessor of Descriptive

linguistics?By whom were Franz Boas' ideas later developed?What method did the Descriptive linguistics criticise and what ap-

proach to phonemes did it propose?Which of the two methods – the oppositional or the distributional –

gives a more exact description of the linguistic facts?3. Who were the founders of Descriptive linguistics?What was the role played by E. Sapir in Descriptive linguistics?How can we characterise L. Bloomfield's works?By whom were Bloomfield's ideas later developed?4. What are Bloomfield's main ideas?How did Bloomfield understand language?What is the role of meaning in Bloomfield's theory? Can you give

the definition of meaning in accordance with Bloomfield's theory?What is the grammatical meaning of an utterance?How does Bloomfield understand grammar?

51

5. What is the definition of the sentence given by Bloomfield?What is the favourite sentence-type in English according to Bloomfield?What are the main positions of the ACTOR-ACTION sentence-type ?How are the main form-classes defined?6. What is the new approach to the breaking up of the word-stock

into classes of words proposed by Bloomfield?How are the form-classes of a language most easily described?How are the long form-classes further subdivided?What do the symbols N and V represent?How do we understand the symbols NP and VP?What is the furthest division of the N-class and the V-class?What is the result of the selection of the N and V subclasses?By whom was the division of the word-stock into form-classes fur-

ther developed?By whom was the concept of utterance introduced? What prompted

the introduction of this concept?7. What is the theory of the Immediate Constituents? Who was the first to introduce the IC? Who can we suppose might

have influenced the idea of the IC?

1.8. TRANSFORMATIONAL-GENERATIVE GRAMMAR

From the transformational method there was only one step to thecreation of a new type of grammar, viz. transformational-generative gram-mar. This method and the method of ICs had paved the way for the emer-gence of a grammar that could account for the generation of the sen-tence.

Unlike the structural grammarian, the transformational-generativegrammarian is not content with describing what he finds in a corpus ofsentences collected from native speakers. He is interested in possiblesentences, i.e. the speaker’s-hearer’s knowledge of a language (compe-tence), rather than in his actual use of it (performance).

There are two types of transformational-generative grammar:1) the Harris – Chomsky grammar and2) the Chomsky grammar.The first type of transformational-generative grammar (TG) was

developed by Harris with his pupil Chomsky, in the 1950s. By the end ofthe decade their paths had separated.

But first let us discuss the model of TG as worked out by Harris inassociation with Chomsky. According to this model, a language consists

52

of a limited number of kernel sentences (i.e. structurally the most simplesentences) and their transforms, i.e. structures derived from them. Ker-nel sentences are generated by the use of the IC model. The set of rulesshowing how a sentence is generated is called rewrite rules, or rewritingrules. Consider the kernel sentence The man hit the ball. This sentence isgenerated by the application of the following rules:

1) Sentence – NP + VP2) NP – T (a determiner) + N3) T – the4) N – man5) VP – V + NP6) V – hit7) NP – T + N8) T – the9) N – ballThis sentence is derived by the use of 6 rules (rules 7, 8, 9 are

recursive, i.e. they have already been used before). From this sentence,applying transformational rules, we can derive other sentences, such asThe ball was hit by the man; Did the man hit the ball?; The man did nothit the ball; What the man did was hit the ball; It was the man who hitthe ball, etc.

The principal transformational rules that can be applied to kernelsentences include:

1) expansion of the verb phrase and the noun phrase, e.g. John is athome. – John must be at home.

We like him. – We came to like him.John is walking. – John is walking in the park.The verb in the kernel sentence can be expanded by using modal

and aspective verbs; the noun by restrictors (articles, pronouns), e.g.John is at home – The John (i.e. our John) is at home.– Poor John is at home.– Mary’s John is at home.2) permutation – change of the word-order, e.g.He is a student. – Is he a student?Jane sent me a letter. – Jane sent a letter to me.3) introduction of functional words, e.g.He arrived tonight. – Did he arrive tonight?Ted is clever – How clever Ted is!4) use of introducers (there, it), e.g.A bell rang. – There rang a bell.5) deletion of an element, e.g.Would you like a cup of tea? – A cup of tea?6) use of negation words, e.g.

53

The evening was warm. – The evening was not warm.7) passivisation, e.g.The teacher praised the boy. – The boy was praised by the teacher.Kernel sentences can be nominalized, i.e. they can be transformed

into noun-phrases (NP) which preserve the semantic relations of the ker-nel sentence, e.g.

The bird sings –1) the singing of the bird;2) the song of the bird;3) the bird’s song;4) a singing bird.To sum up, this model of TG is divided into three parts: 1) phrase-struc-

ture rules, 2) lexicon, and 3) transformational rules. First we begin with thephrase structure rule which says: S – NP + VP. Then we select the rules that areused to generate NP and VP. Then we turn to the lexicon and substitute wordsfor the symbols. Having thus generated a kernel sentence, we can now deriveother structures by using appropriate transformations. This model of TG israther ‘democratic’: it does not require that the transformations should fullypreserve the meaning of the kernels – they may or may not preserve it. Besides,it is very simple. Hence its great popularity among teachers of English.

As already mentioned, the second type of TG was worked out by N.Chomsky, who radically moved away from the first type by distinguish-ing two levels of the sentence – surface and deep. Besides, Chomskygave up the concept of kernel sentence – his model aimed to show howall sentences (simple and composite) are generated in English [3, p. 128].

So what is surface structure and what is deep structure? We will beginwith deep structure. A deep structure is a structure generated only by phrase-structure rules and lexical rules, e.g. not John past can sing well. A deepstructure that has been transformed into a grammatical English sentence iscalled a surface structure, e.g. John could not sing well. All grammaticalEnglish sentences are surface structures; underlying each one is a deep struc-ture. The deep structure of a sentence is a kind of ‘springboard’ for otherstructures which are generated by the application of transformational rules.

As compared to the first type of TG, the Chomskyan TG imposesone important restriction on the transformations applicable to a deep struc-ture, viz. the transformations must not change the meaning of the deepstructure. In the first type of TG, Harris and Chomsky would derive, forinstance, yes/no questions from related declaratives:

Tom is sick.– Is Tom sick?He heard us. – Did he hear us?But as the transformation would change the meaning of the sen-

tences, questions are not derived from declaratives. The idea of interro-gation must be presented in the deep structure of a question, e.g. Q (ques-

54

tion) + Tom + present be + sick. This goes to say that declaratives andquestions are based on different deep underlying structures.

What is especially new and useful in this type of TG is the observa-tion that grammar is a device for generating grammatical sentences. Therules comprising this grammar are limited in number, but the sentenceswe generate by means of those rules are infinite. Although most of thesentences we encounter every day are totally new to us, we have nodifficulty understanding them because the rules they are based on arevery well known to us. A good knowledge of the rules enables the speak-er to ‘create’ new sentences every time he speaks a language.

Special mention should be made of the importance of the conceptdeep structure. TG grammarians would agree that this concept helps usto account for ambiguity and predict it, e.g. Flying planes can be danger-ous. The sentence Flying planes can be dangerous is ambiguous becauseit can be related to different deep structures. It will be obvious that theambiguity of the sentence Flying planes can be dangerous as well as theambiguity of other sentences can also be accounted for using the firsttype of TG, viz. using the concept of kernel sentences. Consider, for achange, the sentence Hunting tigers can be dangerous: The sentenceHunting tigers can be dangerous is ambiguous because it can be treatedas deriving from two types of kernel sentences (A and B).

Harris’s model is more in keeping with the principles of structuralgrammar (the emphasis is on structure) than Chomsky’s (the emphasis ison the rules used to generate linguistic structures).

Other transformational-generative grammarians, such as Charles Fill-more, soon came to see a major inadequacy of the Chomskyan model whenthey had to account for such sentences as John bought the book from Maryand Mary sold the book to John. Semantically, the two sentences describethe same situation, but since they do not contain the same content words (i.e.nouns, verbs, adjectives, and adverbs), we cannot derive them from the samedeep structure. Therefore, they suggested abandoning the Chomskyan deepstructure for a more abstract deep, or semantic, structure, which could allowus to generate the said structures. Such a deep structure is Agent + Process +Affected + Recipient. Mary is the Agent, sold is the Process, the book is theAffected, and John is the Recipient. Given this semantic structure, we canderive either John bought the book from Mary or Mary sold the book toJohn. The use of one or the other construction is the speaker’s choice: if thespeaker wishes to use John as the Theme (or the point of departure), he willgive preference to the first pattern, and if the speaker wishes to use Mary asthe Theme, he will give preference to the second pattern.

The type of grammar which is concerned with the generation ofsemantic structure and derives linguistic structures from it is called gen-erative semantics.

55

The era of structural and transformational-generative grammar hasalready come to an end. However, their achievements have not vanishedwithout trace: they have been incorporated into present-day traditionalgrammar.

1.9. CONTEMPORARY DESCRIPTIVE LINGUISTICS

The main contribution of the American Descriptive School to modernlinguistics is the elaboration of the techniques of linguistic analysis. Themain methods are the Distributional method and the method of ImmediateConstituents. The scientists belonging to this branch of linguisticsunderstand language as one of the semeiotic systems, that is a system ofsignals by which people communicate. There are other semeiotic systemsbesides the natural languages: 'gesture language', Morse, consisting of dotsand dashes, traffic lights, and others. Animals are supposed to have semeioticsystems of their own. This problem is now under investigation. The vocalor natural language is the most important of all the semeiotic systems usedby people. A natural language is a system of vocal signals. These signalsare arbitrary in the sense that they are not inherent or anyhow connectedwith the nature of things they refer to. Every human being learns the systemof the language of his community (aggregation) and by and by he begins tounderstand what people around him say and then begins to speak himself.The task of a scientist is to observe and to describe how people actually saythings, but he should not prescribe how things should be said.

The research is carried out in the synchronic plane. Languages arestudied as spoken languages only, the point of view of their historicaldevelopment is utterly neglected for the time being.

No comparative studies are carried out, only one language, the givenlanguage is being studied (this is called 'the mono-lingual approach'). Thisprinciple has been the basic progressive feature of Descriptive linguistics;it enabled the Descriptive linguists to do away with the traditional approachthat made the scholars understand any language through the norms of Latingrammar, thus distorting the peculiar structure of the language studied.

The most important part of the study is 'field-work' that comprises threeparts: (1) the work with native informants (people who speak the languagestudied by the linguist as their mother-tongue); (2) the filing of results (dif-ferent systems of indexes and slips are used); (3) systematisation.

The writing systems also receive considerable attention. This branchof linguistics is called 'Graphemes' or 'Graphonomy'.

When the methods of Descriptive linguistics were extended to thestudy of such languages as English and other languages well known tothe linguists, the analysis was continued in such a way as if the language

56

were unknown and the linguist was to decipher it as if it were the crack-ing of a code.

Descriptive theory recognises the following fundamental conceptsfor analysing linguistic material:

Utterance: "An utterance is any stretch of talk, by one person, be-fore and after which there is silence on the part of the person."

Sentence: The definition given by Bloomfield is accepted.Structural meaning: The structural meaning of a sentence (the mean-

ing signalled by the parts of the sentence (or form-classes) irrespectiveof their lexical meanings.

The ideas of structural meaning of a sentence-structure was introducednot only by L. Bloomfield, but also by Acad. L. V. Shcherba who gave hisfamous example of the structural meaning in the non-sensical sentence«Глотая куздра штеко будланула бокра и куздрячит бокренка ».

Environment or position of an element is understood as a set of theneighbouring elements.

Distribution: The distribution of an element is the total of all envi-ronments in which it occurs. There is a second definition used by De-scriptive linguistics, namely: Distribution is the class of elements thatoccur in the same position (in the same environment).

Contrastive distribution is understood as a difference of two lin-guistic units occurring in the same environment and changing one lin-guistic form into another linguistic form, e.g. the zero suffix as againstthe -s suffix: pen – pens, book – books, etc.

Non-contrastive distribution is understood as a difference of twolinguistic units occurring in the same environment without changing onelinguistic form into another linguistic form, e.g. hoofs – hooves etc.

Complementary distribution: Two units are said to be in complemen-tary distribution if only one of them normally occurs in certain environ-ments and only the other normally occurs- in other surroundings, e.g. -(e)s[z], [is], [iz] in 'rooms', 'books', 'boxes', etc.

Morpheme: The morpheme is a linguistic form which bears no par-tial phonetic-semantic resemblance to any other form.

Allomorph: An allomorph is a variant of a morpheme which occursin certain environments. Thus a morpheme is a group of one or moreallomorphs (or morphs).

The allomorphs of one and the same morpheme(1)must be in complementary distribution;(2)the sum of their environments must be equal to the sum of envi-

ronments of some single morpheme in the language, e.g. the allomorphs[z], [is], [iz] together have the same set of environments as the singlezero suffix: room – rooms, book– books, box – boxes, etc.;

(3)they must be similar in meaning.

57

Prof. A. Smirnitzky, who recognised the above mentioned criteria,pointed out that the basic phonetic variants of the free allomorphs [z], [s],[iz] and [d], [t], [id] are respectively [z ] and [d] as they are produced bynative speakers in conditions where any variant could appear after vowels;the voiceless variants [s], [t] are the result of assimilation to the precedingvoiceless consonants, and [iz], [id] represent the historically earlier formswith an [i] preserved between two phonemes of a similar character.

We may note that the differences in the environments are the samefor such series of pairs as:

boy – boys, pay – paidbook – books, talk – talkedman – men, take – tooksheep – sheep, cut – cutThe same criteria hold for identifying suppletive forms. According

to Prof. A. Smirnitzky, the paradigm of the verb 'go' is recognised on thebasis of three features (when compared to the paradigm of such a verb as'look': (1) the lexical meaning of the forms is identical, (2) they are incomplementary distribution, (3) the set of environments of suppletiveforms is the same as that of the suffixed forms, namely:

go – look – went looked gone – looked.Instances of complementary distribution are numerous in morphol-

ogy. Forms with phonemic replacement and suppletive forms can be de-scribed in terms of complementary distribution on the basis of patterncongruity, e.g.

book – books, look – lookedclass – classes, start – startedman – men, write – wrote, go – went.Form-classes or positional classes: In Structural linguistics this classi-

fication is set up on the basis of a particular choice of diagnostic co-occur-rents: 'cloth' and 'paper' both occur, say, in "There is" where 'diminish' doesnot appear: we call this class N. And 'diminish' and 'grow' both occur, say, in"It will –" where 'paper' and 'cloth' do not: we call this class V. The diagnosticenvironments are chosen in such a way that the resulting classes permit com-pact statements about co-occurrence. For example, 'cloth', 'paper', 'diminish','grow' all show some differences in their environments, so that a simple sum-mary can be made. In terms of the classes N and V we can say that every Noccurs before some V in the environment 'the – V, and every V occurs in theenvironment 'the N –' for some N. (See Z. S. Harris [10, p. 284]).

Construction: A construction is any significant group of words ormorphemes.

Constituent: A constituent is a linguistic form that enters into somelarger construction, e.g. the constituents "the old man" and "has gone tohis son's house" constitute a sentence.

58

Immediate constituent: An immediate constituent is one of the twoconstituents of which the given linguistic form is directly built up. Endo-centric construction: Endocentric constructions are of two kinds: co-or-dinate and subordinate; they have the following distinctive feature incommon: the position of the construction in the sentence is the same asthe position of one of its constituents.

Exocentric constructions differ from endocentric constructions inthat they have a position (or function) different; from the position ofeither of their constituents.

Linguistic levels: The main units (elements) of language are usually rec-ognised by Descriptive linguistics: the phoneme and the morpheme. A thirdlevel is often recognised, the level of constructions or the syntactic level.

Any utterance or part of utterance can be described in terms of mor-phemes and any morpheme can be described in terms of phonemes. Thusany utterance can be presented on the phonemic level (as a sequence ofphonemes) and on the morphemic level (as a sequence of morphemes).

The notion of levels is closely connected with that of isomorphism.Isomorphism means similarity (or parallelism) of relations between theunits concerned.

The structure of language is understood as consisting of differentlevels connected with each other by the relation of hierarchy. Hierarchymeans that the units of a lower level are elements of which elements of ahigher level are built up (composed) and into which they are analysable.

1.10. THE EXPLANATORY POWER OF NON-STRUCTURALDESCRIPTIVE, STRUCTURAL DESCRIPTIVE

AND TRANSFORMATIONAL-GENERATIVE GRAMMAR

Grammar should seek to explain how language is structured, whatfunctions its structures perform, what rules are used to form sentences orword combinations.

Different grammars solve these problems with a different degree ofsuccess. We will test the explanatory power of each type of grammar byconsidering the sentence, the largest unit of grammar over which a ruleof grammar can operate.

1.10.1. Non-Structural Descriptive GrammarTraditionally, the sentence is a group of words that expresses a complete

thought or a group of words that contains a Subject and a Predicate. The firstdefinition, which is a notional one, is rather subjective since there are no crite-

59

ria by which we can judge the completeness of a thought. The second defini-tion is not satisfactory either because it rules out verbless sentences.

The sentence is a unit of communication, which suggests that anystructure that can perform this function is a sentence. (This interpreta-tion of the sentence is also traditional in the sense that it is neither struc-tural nor transformational-generative; it is present-day, or contemporary,traditional.) To be more precise, any structure that can express new in-formation is a sentence.

Traditional grammarians devote a great deal of time and energy tosentence analysis. According to the traditional method, the sentence isanalyzed in terms of the parts of the sentence: Subject, Predicate (princi-ple parts), Object, Attribute, Adverbial Modifier (secondary parts).

Having identified the parts, traditional grammarians proceed to charac-terize them morphologically: What part of speech is it? In what form, tense,aspect, mood, voice, etc. is it expressed? The main shortcoming of the tradi-tional method is that sentence analysis is based on syntactic notions whichare not defined clearly. Such being the case, syntactic analysis turns into anart: one and the same constituent is often given more than one analysis.

Consider the sentence: John wants to go there. There being no clearcriteria for distinguishing parts of the sentence, we cannot say for certainwhether the infinitive is part of the verbal predicate or the object.

Consider another sentence: He was known to like her. It is not clearwhether the Subject is only he or he + to like her. Both analyses can befound in traditional grammar. A similar situation can be observed when weanalyze secondary parts of the sentence, e.g. He swam across the river. vs.He swam the river. The word the river is often given the same analysisdespite a difference in pattern. The same indeterminacy concerns the anal-ysis of a key in He opened the door with a key: is it an Adverbial Modifierof Manner or a Prepositional Object? All this suggests that traditional sen-tence analysis is endowed with serious problems.

Traditional grammarians cannot adequately cope with ambiguity inlanguage, e.g. He is a man to watch. They are aware of the ambiguity andsay that the sentence is ambiguous because a man may be given twointerpretations: subjective and objective – He is a man who watches orHe is a man who is watched.

Although it is a correct account, we cannot say it is adequate: itdoes not say what is "responsible" for the ambiguity.

A similar problem arises when we analyze the sentence He fed herdog biscuits. The traditionalist will be forced to say that the sentence isambiguous, or that it is a trick sentence. He/ she will say that her may betreated as the Indirect Object of fed (fed her) or the possessive restrictor(determiner) of the noun dog (her dog). But he/she will not be in a positionto answer the question what makes it ambiguous. Also, in treating the struc-

60

ture the love of God, the traditionalist will admit that it is ambiguous, forthe constituent God may be given a subjective and an objective interpreta-tion. Although the traditional analysis is correct, it is not adequate.

Traditional grammarians treat syntactic structures as independentunits, although they are aware of existing derivational relationships be-tween them. For instance, such relationships are assumed to exist be-tween active and passive sentences, between simple and composite (com-pound and complex) sentences, between declarative, negative, interrog-ative, and exclamatory sentences.

However, the existing relationships are not formalized in terms ofparadigmatic relations. Traditional grammarians do not see such rela-tionships between predicative structures and non-predicative ones, e.g.John arrived vs. John’s arrival.

It will be obvious that what has been said about traditional grammarso far is only true of the type of traditional grammar which existed beforethe advent of structural descriptive grammar.

These days traditional grammar, which continues to be based onmeaning and function, incorporates the achievements of the past and thepresent, and, as in the past, is used as a reference source by teachers ofEnglish and as a point of departure by scholars. The adoption of newmethods of analysis (structural, statistical) greatly enhances its explana-tory power.

1.10.2. Structural Descriptive GrammarStructural grammarians prided themselves on being true linguists:

they based their analysis on actual English, giving preference to spokenEnglish; they used 'discovery procedures' such as distribution and substi-tution tests, transformations of various sorts, etc. As compared to tradi-tionalists, they were more analytic: their attention was on segmentationand categorization (i.e. labeling structures).

Rejecting traditional concepts and methods as unscientific, struc-tural grammarians focused on the development of a grammar which wouldbe devoid of 'old illnesses', a grammar not influenced by Latin or Greek.It was in the field of syntax that Latinization was the most obvious.

As already indicated, structural grammarians put forward a newmethod of sentence analysis, viz. the immediate constituent (IC) method.The essence of the method is that the sentence is viewed as being com-posed of layers, or levels – higher and lower. The layers are subordinateto each other. By means of this method we can identify the syntacticrelations between constituents that are adjacent (next) to each other. Theterm immediate means that there is no other syntactic element in be-tween. Consider: Mary married John.

61

As can be seen, the sentence is divided into two immediate constit-uents Mary married John. The difference between the two sentencesconcerns the relationship between the man and the adverbial construc-tions in the red cap and in the right arm with respect to the verb shot: inthe first sentence the second cut is between shot and the man while in thesecond sentence it is between the man and the right arm.

In traditional terms, in the first sentence in the red cap is an At-tribute to the man while in the second sentence in the right arm is anAdverbial of Place. So far so good. But how can we show different struc-ture in the case of the love of God?

Traditionalists would say that God may be treated in two ways: itmay have a subjective and an objective function. The structuralists’ im-mediate constituent method is powerless here. However, the transforma-tional method copes with the ambiguity easily: The love of God - 1. Xloves God. 2. God loves X.

Returning to IC analysis, we must say that it only identifies syntac-tic relations, or dependencies: it does not fill the relations with content.Traditional sentence analysis into sentence parts, or into the syntacticfunctions of the sentence constituents, seems to be more acceptable sinceit does not ignore syntactic function.

Structuralists rejected the traditional method of the classification ofwords into word-classes and replaced it by the distributional method, or,roughly speaking, the positional method. As there are few forms in En-glish, the behavior of a word in the context becomes a crucial factor inclassifying words. But the distribution of a word is practically the sameas the function of a word in a sentence. This suggests that the traditional-ist also makes use of the same principle as the structuralist. Despite thesimilarities, structural grammar has an advantage over traditional gram-mar in being more rigorous as concerns linguistic analysis.

Giving an overall evaluation of structural grammar, it is necessaryto point out that it pays special attention to analysis, to the distinction ofstructural units (phonemes, morphemes, lexemes, sentences). Structural-ists were too preoccupied with the sequence of phoneme-to-sentence andfailed to see the interrelationships outside the sentence.

1.10.3. Transformational-Generative GrammarTransformational-generative grammar does not teach us how to anal-

yse sentences; it teaches us how sentences are generated in a language.Neither traditional nor structural grammar was interested in the genera-tion of sentences.

What is more, the recognition of two types of structure – surface anddeep – makes it possible to relate all the sentences of a language and even

62

different languages: sentences and languages which are quite different onthe surface often show many similar features in their deep structures.

Transformational-generative grammar can account for any structur-al ambiguity by relating ambiguous constructions to two (or more thantwo) deep structures. Ambiguity is the result of the neutralization of thedeep, or underlying, relations. Consider:

Hunting tigers can be dangerous.This sentence can be related to two different deep structures:DS (1) Tigers + pres. hunt + X + Tigers pres. can be + dangerous;DS (1) X pres. hunt + tigers + It + pres. can be + dangerous.As already mentioned, TG makes it possible to relate one sentence to

another: sentences are related if they derive from the same deep structure:DS The manager + past write + the letter-The manager wrote the

letter. The letter was written by the manager.Besides, TG can relate sentences to other structures: the structures

The letter written by the manager and the manager’s having written theletter are related through the same deep structure – The manager pastwrite + the letter.

However, if we apply the Harris model, we shall be able to derivemore structures from the same deep structure, for the Harris model, incontrast to the Chomskyan model, is ‘more democratic’ – it is not boundby the requirement that transformations should not change the meaningof the transforms. Thus the Harris model will derive all the structuresderived by the Chomskyan model and others:

Did the manager write the letter? The manager did not write theletter. Who wrote the letter?

What did the manager write? The writing of the letter by the manager.For the manager to write the letter. Because the manager wrote the letter, etc.

In view of this, the Harris model is more powerful: it can derivemore structures from the kernel sentence. Besides, it is more simple.Being more simple, it is easier to use in the classroom. Transformationsdemonstrate the cohesiveness of language where simpler constructionsare built into more and more intricate ones.

The shortcoming of TG lies in its complexity. Besides, language ismore complex than transformational grammarians thought it was: it con-tains structures that can only be described by a very sophisticated (intri-cate) formal apparatus which would render it useless in the classroom.

Transformational grammar concentrates on competence and ignoresperformance, i.e. the actual use of linguistic structures, which suggeststhat the picture of a language presented by TG is one-sided.

63

2. GRAMMAR: THEORETICAL ISSUES

2.1. PRACTICAL AND THEORETICAL GRAMMAR

The actual definition of grammar is determined by pragmatic factors. Ifwe wish to learn to speak and write, we will focus on the system of rules thatunderlie a given language, and if we wish to describe the structure of a lan-guage, we will focus on the units that make up the language and their rela-tions, and if we wish to understand how speakers of a given language pro-duce and understand sentences, we will focus on the nature of the rules used.

Hence we can speak of two types of grammar: practical and theoreti-cal. Practical grammar gives practical rules of the use of the linguistic struc-tures while theoretical grammar gives an analysis of the structures in thelight of general principles of linguistics and the existing schools and ap-proaches. The aim of theoretical grammar of language is to present a theoret-ical description of its grammatical system. To achieve this aim it is necessaryto scientifically analyze and define its categories and study the mechanismsof grammatical formation of utterances in the process of speech production.

Modern linguistics is essentially based on the systemic conceptionof language. System in general is defined as a structured set of elementsrelated to one another by a common function. The interpretation of lan-guage as a system develops a number of notions, namely: the notions oflanguage levels and language units, paradigmatic and syntagmatic rela-tions, the notions of form and meaning (function), of synchrony and di-achrony, of analysis and synthesis, and some others.

The principle adhered to here will be that of functional grammarwhose main goal is to reveal the properties of the units of grammar bothin their associative paradigmatic) and linear (syntagmatic) relations,andto give a clear idea of the structural essentials of the particular languageunder study with all its surface and deep seated regularities, in otherwords, to describe the language as a system and its constituents as partof that system, or of a microsystem, within an organic whole.

2.2. PARADIGMATIC AND SYNTAGMATIC RELATIONS

Lingual units stand to one another in two fundamental types of rela-tions: syntagmatic and paradigmatic.

Syntagmatic relations are immediate linear relations between unitsin a segmental sequence (string).

64

One of the basic notions in the syntagmatic analysis is the notion ofsyntactic syntagma. A "syntactic syntagma" is the combination of wordsor word-groups one of which is modified by the other. To syntagmaticrelations are opposed paradigmatic relations. They exist between ele-ments of the system outside the strings in which they co-occur. Theseintrasystemic relations find their expression in the fact that each lingualunit is included in a set or series of connec tions based on different for-mal and functional properties.

Paradigmatic and syntagmatic relations are not isolated from oneanother. Paradigmatic relations co-exist with syntagmatic relations in sucha way that some sort of syntagmatic connection is necessary for the real-ization of any paradigmatic series. This is revealed to the full in a classi-cal grammatical paradigm. It presents a productive series of forms. Aparadigmatic form – a constituent of a paradigm – consists of a stem anda specific element (inflexion, suffix, auxiliary word). The function of agrammatical paradigm is to express a categorial meaning.

Paradigmatic relations are associative. This means that they unite sim-ilar units on one paradigmatic axis to form a paradigm or set in which unitsrelate to each other by association with some distinctive feature (DF), orcategory, or kind of relationship common to all members of such a para-digmatic set. E.g.: grammatical (form-building) morphemes of case (wherethey exist), tense, person-number form a paradigm the members of whichare associated on the ground of the grammatical category (or DF) of case,tense, etc. Accordingly, grammemes that carry a certain grammatical infor-mation are united in a paradigm on the basis of some category or DF. E.g.:grammemes (filled with linguistic material): t) (I) am speaking, (he) isspeaking, etc. or 2) speak – spoke – will speak – would speak are similarunits with a common general categorial meaning that brings them togetherinto a paradigm in which there is some ground for comparison common toall its members; thus in (1) this common ground is the grammatical mean-ing of duration (the distinction lies in the person); in (2) it is the generalmeaning of tense (time) that unites the four grammemes.

The units compared belong to the same rank and they cannot be puttogether on a syntagmatic chain to form a communicative entity (such asentence, as I am speaking, he is speaking, they are speaking ... would benonsensical).

Words united in one part of speech also stand to each other in para-digmatic relations, as they are similar units (all nouns put ogether, verbs,etc.) having a common general grammatical meaning (of substance, ac-tion, etc.). Paradigmatic relations are associative because they are basedon an identical grammatical meaning of the unit. Since meaning is em-

65

bodied in a certain form, a paradigm may be marked by form-building(morpheme) alternations.

In as much as grammatical meaning cannot be directly observed, it canbe reached at only through the close study of the interrelations among gram-matical units, since the abstract grammatical meaning is ways relational.

Associative paradigmatic relations are clearly cut with morpholog-ical units represented by one and the same part of speech. MorphologicalUnits are by nature paradigmatic; therefore with them paradigmatic anal-ysis comes first. The paradigmatics of morphological units (morphemes,grammemes, and parts of speech) constitutes the framework of a lan-guage, reflects, both in form and meaning, the properties inherent in thatparticular language, composes what may be called the morphologicalsystem of the language.

Matters are made more complicated with the establishment of para-digmatic relations among syntactic units which consist of no less thantwo words (with a few exceptions of clipped constructions and sentencemembers). They are syntagmatic by nature, since they represent a linearstructure. With them syntagmatic analysis comes first, the more so thatdifferent syntactic combinations and structures are but abstract modelswhich have to be singled out on the basis of a preliminary syntagmaticanalysis of linear structures existing in that particular language.

The paradigmatics of syntactic units presents generalizations of typesof groups and sentences into a set of patterns and communication types, andrelations existing among them. Thus, paradigmatics of syntactic units is themodelling of syntactic structures and syntactic relations in a language.

Syntagmatic relations expose linear relationships of dissimilar unitsfollowing each other in the syntagmatic chain. For instance, in the sen-tence I am talking about grammar different words and forms stand to theunit preceding or following them in syntagmatic relations. Such rela-tions are obvious, seen on the surface,one can observe them in speech ora text and generalize them on the basis of common knowledge of thelanguage under study.

Syntagmatic relations of a morpheme are revealed in its combinato-rial potential with certain types of words or bases. The syntagmatic rela-tions of a grammeme expose what is usually called the use of grammati-cal forms in speech or text. Syntagmatic relations of wordclasses (partsof speech) reflect their combinability with each other, their potency toenter certain combinations (groups). Syntagmatic relations of parts ofspeech produce word-groups – the next unit on the rank scale of gram-matical units and the first unit of syntax. The ultimate constituent of agroup or sentence, in other words, the marginal element of syntactic unitswhich does not allow further cuts into smaller units is the word. The

66

group and sentence (clause) are composed on the syntagmatic basis, be-ing combinations of no less than two words,and any description of suchunits in the text is first of all syntagmatic.

Outside the border of a purely syntactic analysis of a sentence areleft different stylistic, pragmatic functions of grammatical units, mostlyof syntactic structures. These present a separate field of study, whichrequires the introduction of a variety of additional notions that outstepthe limit of morphology.

Paradigmatic and syntagmatic relations taken together cover all sidesof a linguistic sign. They are studied by different aspects of grammar.

Generally speaking paradigm is interpreted as a formal way of say-ing of a type of something: a pattern, a model; in grammar it means a setof all the differnt forms of aword, its wordforms, thus verbs are conju-gated and nouns are declined, adjectives and adverbs have forms of de-grees of comparison, their forms correspondingly make the verbal, nou-nal, adjectival or adverbial paradigms.

Etymologically paradigm is a Greek word, para – beside, and deig-ma from Gr. deiknumi, ‘to show’ an example of a word in its variousinflexions.

Paradigmatic (associative) and syntagmatic (linear) relations ofgrammatical units in the system of a language are studied within twosections of grammar – morphology and syntax. It does not mean, howev-er, that paradigmatic relations are studied in morphology, and syntag-matic, in syntax.

The specific properties of morphological and syntactic units callfor a definite distribution of morphological and syntactic studies. How-ever, there is a great variety of controversial views on the morphology –syntax division in grammar. These views depend on the general theoret-ical outlook of linguists of different schools.

For instance, the representatives of American descriptive linguis-tics (Z.Harris, Ch.Hockett, E.Nida, A.Hill and others) have always dis-carded this division as unnecessary.

It was quite a natural outcome of the general postulates of that school,the advocates of which departed from the distribution and arrangement ofmorphemes in linear succession, in speech (or text). The morpheme wasproclaimed as the main speech unit. In other words, the initial point ofstudy was the syntagmatic combination of linguistic units, directly observ-able, and not their paradigmatic, systemic properties lurking behind theobvious. The linguists of the descriptive trend define the morphologicalstructure of words and the parts of speech by means of substitution.

If the morphemes (including isolated words which are regarded asmorphemes) occupy the same position in a syntagmatic chain, they are

67

said to belong to the same class. Thus classes of units are detected onlyin a linear structure by means of substitution. They are called equiva-lence classes.

With such an approach to the detection of linguistic units, theirmeaning, their associative relations to each other, their grammatical cat-egories fall out of the sphere of analysis and there can be no question ofthe hierarchy of morphological and syntactic units; therefore the mor-phology – syntax (M – S) subdivision of grammar simply cannot exist.

The later trend in grammar in general, and English grammar in par-ticular, the transformational grammar (TG), concentrated on tactic struc-tures. In a great number of works based on TG written in different coun-tries, numerous manipulations are performed on word combinations ofvarious length, and here, naturally, one cannot expect the appearance ofthe distinction between morphology and syntax.

James Muir in his A Modern Approach to English Grammar. AnIntroduction to Systemic Grammar (London, 1972) based on a principledifferent from those discussed above strives to present the grammaticalstructure of English as a systemic whole (which is evident from the title).Here, however, the terms 'morphology' and 'syntax' as the frames for thedescription of language units are absent. Instead the notions of SurfaceGrammar and Deep Grammar are introduced which cover the directlyobservable facts lying on the surface and associative, deep relations notobvious at first sight. However, these notions do not stand to the notionsof morphology and syntax in one to one correspondence.

Speaking of the well-known authors of English grammar of the oldergeneration, one must state that here the M – S division undergoes differenttreatments. For example, G.O. Curme introduces such notions as 'syntaxof the noun, verb, etc' and deliberately confuses morphological and syntac-tic phenomena. Case, according to him, can be expressed by a prepositionplus noun (of the boy, to the boy) or by position: e.g. in / saw the boy thenoun stands in the accusative case. But both positions in a sentence andword combinations are, to be sure, syntactic notions, and not morphologi-cal. Thus the M – S boundary remains unspecified, vague.

A prominent linguist and the author of books on English grammar,O.Jespersen treats the function and meaning of all morphological gram-matical categories in syntax (case, tense, mood, etc.). Morphology withhim is limited to word-building. This happens because he departs fromthe assumption that all facts of language can be defined only in combina-tion with each other, and, therefore, it is mostly syntactic relations thatmatter in the detection and recognition of language facts.

It is evident from the aforesaid that the solution of this problemdepends on the general linguistic background of the author or/and the

68

trend he belongs to. If a functional approach to the grammatical structureof the language under study is carried out, then the M – S division isbrought into being.

It remains now to specify what functions of English morphologicaland syntactic units, situated within the limit of a sentence, come withinthe province of M and S.

The sphere of English morphology, starting from the lowest gram-matical unit on the rank scale, comprises the identification of formbuildingmorphemes and their grammatical functions (meaning). The next step con-sists in establishing the inventory of verb grammemes, both synthetic andanalytic by structure, which build up the paradigmatic macrosystem of theEnglish verb. As there is no developed declension in nouns (and none inadjectives) the corpus of grammemes – exponents of grammatical catego-ries – is concentrated within the verb only. It follows from our general goalthat the main object here is the associative (relational) grammatical mean-ings of grammemes which correlate within the micro- and macrosystemsof the verb, and manifest themselves in corresponding forms (grammemes).

The part of morphology which deals with inventories of morphemes,grammemes, and parts of speech, i.e. with the paradigmatic setup andrelations within the noun and the verb (no other word-class possessesform-building exponents, the markers of degrees of comparison in adjec-tives and adverbs are of word-building nature) may be termed paradig-matic morphology as opposed to syntagmatic morphology which dealswith the usage of morphological units, the actual employment of formsin linear succession, in speech (text). The study of the conditions underwhich grammatical forms are used, the items of the immediate environ-ment which affect the meaning and employment of a morphological unit,all come within the sphere of syntagmatic morphology.

It deals with the combinability of morphological units and formsfrom the standpoint of the realization or neutralization of morphologicaloppositions, of the grammatical meaning inherent in certain forms. Itbecomes evident, therefore, that 'syntagmatic' is not equivalent to 'syn-tactic'. However, once syntagmatic morphology has to do with combin-ability, it would be natural to expect that there may exist some caseswhen to draw a fast and sharp line between the two notions would not betoo easy, the more so that the main unit of morphology is the word, whichis at the same time the ultimate constituent of a sentence or group.

The word is a two-faced Janus (even many-faced); when it is viewedas the carrier of some generalized, abstract grammatical meaning, it be-longs to morphology, when it is viewed from the point of view of theposition it occupies in relation to other, mostly dissimilar, words (be-longing to different word-classes), it belongs to syntax.

69

Thus, syntax, syntagmatic in its very nature, deals with the positionof words and their mutual relations in word combinations of different length(up to a sentence and beyond it), and, accordingly, with the structure andgeneral meaning of sentence members, word-groups and clauses (sentenc-es). There is no need to break up syntactic studies into two parts, as is donein morphology, because of the specific nature of syntactic units (discussedabove),the study of which always presents a mixture of the identificationof certain abstract models and their concrete, particular realizations.

2.3. LANGUAGE AND SPEECH

The discrimination of language and speech is the fundamental prin-ciple of linguistics. This principle has sustained throughout the wholehistory of the study of language. With a special demonstrative force itwas confirmed by I.A. Beaudoin de Courtenay (end of the XIX c.) andF. de Saussure (beginning of the XX c.) who analyzed the language-speech dichotomy in connection with the problem of identifying the sub-ject of linguistics. The two great scholars emphatically pointed out thedifference between synchrony and diachrony stressing the fact that atany stage of its historical evolution language is a synchronic system ofmeaningful elements, i.e. a system of special signs.

Language in the narrow sense of the word is a system of means ofexpression, while speech is a manifestation of the system of language inthe process of communication. The system of language includes the bodyof material units – sounds, morphemes, words, word-groups, and a set ofregularities or "rules" of the use of these units. Speech comprises boththe act of producing utterances and the utterances themselves, i.e. thetext made up of lingual units of various status.

From the functional point of view all the units of language shouldbe classed into those that are non-meaningful semantically, such as pho-nemes, and those that express a certain semantic meaning, such as words.The non-meaningful units may be referred to as "cortemes", they providea physical cover (acoustic, graphical) for meaningful units; the meaning-ful units, in distinction to cortemes, may be referred to as "signemes".Signeme is a lingual sign. The introduction of a special name for it iscalled upon to show that there is a profound difference between lingualsigns and non-lingual, common signs.

Language and speech are inseparable, they form an organic unity.The stability of this unity is ensured by grammar since it dynamicallyconnects language with speech by categorially determining the processof utterance production.

70

The signeme (lingual sign) in the system of language has only apotential meaning. In speech the potential meaning of the lingual sign is"actualized", in other words, it is made situationally significant as part ofthe grammatically organized text.

The functional dynamics of lingual units in speech is efficientlydemonstrated by the branch of linguistics called "pragmalinguistics".Among other things, pragmalinguistics investigated the relevant contri-bution to the total communicative content of utterances made by differ-ent unit types. In this connection, stretches of speech have been describedthe role of which consists not in the expression of certain meanings, butin maintaining the contact between the communicants, or sustaining the"phatic communion".

2.4. LANGUAGE UNITS AND LANGUAGE LEVELS

Units of language in general, and of grammar in particular, form ahierarchy of interconnected elements, a rank scale. The lowest grammat-ical unit on that scale is the morpheme. Every lower unit forms part of ahigher one. The position of the unit on the low or high step of the rankscale depends on its size. The longer (in linear progression) is the unit,the higher is its place on the linguistic ladder.

Units of language are divided into segmental and suprasegmental.Segmental units consist of phonemes, they form phonemic strings of var-ious status. Suprasegmental units do not exist by themselves but are real-ized with segmental units and express different modificational meaningsreflected on the strings of segmental units.

The segmental units of language form a hierarchy of levels. Unitsof each higher level are formed of units of the immediately lower level.But this hierarchical relation is not reduced to the mechanical composi-tion of larger units from smaller ones, as units of each level are charac-terized by their own, specific, functional properties which provide thebasis for the very recognition of the corresponding language levels.

The lowest level of lingual units is phonemic: it is formed by pho-nemes. The phoneme has no meaning, its function is purely differential.

The second level, located above the phonemic level, is morphemic.The morpheme is the elementary meaningful part of the word built up byphonemes. The morpheme expresses abstract, "significacive", meaning.

The third level is lexemic. Its differential unit is the word. The wordrealizes the function of nomination.

The fourth level is denotemic, its constituent unit is denoteme (no-tional part of the sentence).

71

The fifth level is proposemic. It is built up by sentences. As a sign, thesentence simultaneously fulfils two functions - nominative and predicative.

The sixth level is the level of topicalization, its constituent elementis the "dicteme" ("utterance"). The function of the dicteme is to build upa topical stretch of some text. Being an elementary topical unit of text,the dicteme fulfils four main signemic functions: the functions of nomi-nation, predication, topicalization, and stylization.

1. Morphological unitsa) The MorphemeGrammatical morphemes are scarce in English due to the preva-

lence of analytic or zero ending formations. They are called form-build-ing (morphological) morphemes (e.g.: -ed of the Past) as opposed to word-building (lexical) ones (e.g.: -merit in government, -less in jobless), whichdo not belong to grammar as they serve to create separate lexical sub-classes and therefore lack the wide-range abstraction and universalitycharacteristic of grammatical units. E.g.: -ed covers the whole of theclass of verbs subjected to this kind of form-building (so called regularverbs) whereas word-building morphemes cover a narrow range of wordsgrouped into formal or semantic subclasses.

A form-building morpheme is inevitably a member of an opposition,i.e. is opposed to some other morpheme expressing the same general gram-matical category. It is part of a paradigmatic set of one and the same word;whereas word-building morphemes rarely form an opposition; besides theyare attached to different words with different paradigms.

Grammatical morphemes do not appear isolated; they always formpart of a grammeme (word-form). A morpheme is the only unit which can-not be cut further. As most words which one encounters in an English text(or in speech) are devoid of any grammatical inflexion, it would be uselessfor grammar to label each word or each base to which a form-buildingmorpheme might be added as a <morpheme. We consider it to be morerational, in the process of grammatical description, to call word-forms whichare capable of creating a paradigm and to which some form-building mor-pheme may be added – base, e.g.: speak – base; speaks – base + form-building morpheme -s; speaking – base + -ing, etc. Or: boy – boys – base +-s (plural).

In this connection the singling out of a zero morpheme in Englishmorphology becomes superfluous, as 'zero' means no morpheme at all,and this is too universal a phenomenon in English to call for specialconsideration.

The definition of a morpheme current in linguistics nowadays is thatof 'a minimal meaningful unit' (L. Bloomfield). This definition is defectivein that it does not specify what kind of meaning is understood and it does

72

not mark out the difference between form-building and word-buildingmorphemes. Gleason's definition is more careful: "A morpheme is a mini-mal unit in the plane of expression which relates to some unit of the planeof content". Gleason considers it to be a minimal unit of grammar. Howev-er, morphemes in English are mostly word-building ones, i.e. lexical.

A form-building morpheme may be defined as an element of theword which signals the kind of grammatical meaning attached to it bythe presence of the morpheme. The morpheme itself has a purely rela-tional grammatical meaning which is revealed only by contrast with someother morpheme or grammeme which exposes a contrastive grammaticalmeaning. A morpheme as a unit of grammar is an exponent of a grammat-ical category (or grammatical meaning).

b) The Grammeme (Word-form)The next grammatical unit on the rank scale is the word and its

grammeme (word-form). The term 'grammeme' is convenient because itis monosemantic, its inner form suggests a grammatical unit, and its -emic origin puts it on the same scale as the morpheme and phoneme. Itmay be, though not necessarily is, equivalent to a word, at any rate itpresents an isolated unit, not part of a word. It is a carrier of grammaticalinformation.

When we speak of a word as a grammeme we abstract ourselvesfrom its lexical meaning and concentrate on the kind of grammaticalinformation it carries, e.g.: the grammeme speaks shows the present tense3rd person singular. It can be identified in such a way exclusively due tothe existence of opposed forms, such as speak, spoke, is speaking, etc.contrasted to speaks in different distinctive features (or grammaticalmeanings). Here the relational property of grammatical meaning is re-vealed. The lexical meaning of the word is irrelevant for the detection ofthe type of grammeme.

For practical purposes of grammatical description it is possible touse the term 'word-form' for a concrete lexical unit regarded as a gram-meme, e.g.: worked is a word-form of the verb to work expressing thepast indefinite tense; if viewed as a pattern of some grammatical value ormeaning worked can be said to be a grammeme.

A grammeme may be analytic by structure, which means that it con-sists of more than one word (e.g., has spoken); an analytic grammeme isequivalent to one word on the rank scale as it expresses one unified contentof a word, both from the point of view of grammatical and lexical meaning.

Morphemes and grammemes are directly observable units by na-ture in that they are characterized by a definite material structure of theirown. They can be registered and enumerated in any language, howevercomplicated the system of form-building might be in a language (cf. in

73

Greek and Latin there is a great variety of types of conjugations anddeclensions, but nevertheless they are all registered in grammars). There-fore the system of morphological units is a closed system in that all itsitems are on the surface and can be embraced in an inventory of forms.

Not every word is at the same time a grammeme, but nevertheless itis a unit of grammar as a part of speech. Parts of speech are usuallyconsidered a lexico-grammatical category since, on the one hand, theyshow lexical groupings of words; on the other, these groupings presentgeneralized classes, each with a unified, abstract meaning of its own.The latter makes word-classes a grammatical notion since wide-rangeabstraction is characteristic of grammar. Each part of speech, as a gener-alized word-class possesses a certain valency, i.e. inner potential to com-bine with other word-classes in linear order (in actual speech). In accor-dance with this potency words make combinations (phrases, groups) whichpresent the next unit on the grammatical rank scale.

2. Syntactic unitsIn contradistinction to morphological units, syntactic units – a word-

group (phrase), a sentence member, and a clause (sentence) – do notrepresent observationally accessible, concrete linguistic elements whichcould be studied and differentiated by specific material structure of theirown. They are abstract patterns (models), analogues of certain lexicallyfilled collocations. These patterns are drawn from common linguisticknowledge of men. Word combinations and clauses reflect the higheststage of grammatical abstraction, being pure generalizations of an infi-nite number of concrete phrases and sentences of a language.

Any group consisting of no less than two words allows of a variety ofcombinations, the number of which increases progressively with the in-crease in the number of items constituting a syntactic collocation. The sys-tem of syntactic units is the most open one since no inventory of any con-crete type of phrases can be exhaustive and embrace all the possible ele-ments that might occur in linguistic reality. For instance, in the filling ofthe pattern of the word-group consisting of a transitive verb plus 'direct'object one might postulate certain lexical classes or subclasses of the type(he) bought a book, lifted a weight, knows the language, etc., but even ina special study of the filling of this pattern in a language it would be impos-sible to state that the pattern is closed and no more varieties of the verbaland nominal parts of it can exist in that language.

The higher the unit on the rank scale, the less it is probable to create theinventory of its concrete material structure. It would be absolutely impossi-ble to compile a dictionary of existing word-groups or sentences in a lan-guage, and, actually, no such exhaustive dictionary of word-groups exists.

74

According to the general hierarchy and interconnection of linguis-tic units, it is natural to expect that the word is the ultimate constituent ofsyntactic units; any syntactic analysis beginning with the sentence as aunit, ends with the word, with the part of speech. However, it is rathernot so much the word as a lexical unit or part of speech that is the startingpoint or final touch of syntactic description, as the position of a certainword-form in a sentence (or group) in relation to other words. The word-form may be said to be an elementary syntactic unit from which anysyntactic description departs; however, it is not its lexical meaning or itsmeaning as a morphological form that matters here, but its position andsyntactic function within a node of some meaningful syntactic relation(such as predicative, objective, etc.).

This postulate can be illustrated by a Russian word-form: e.g.: theword-form отца in жду отца, дом отца, according to its position,expresses either objective or attributive (possessive) syntactic relationsand enters groups of different syntactic meaning. Or the word-form read-ing in Reading is pleasant. He stood at the window, reading. Still read-ing, he boarded the train acquires a different syntactic function, and, inthis case, a different morphological meaning. Not every sentence neces-sarily contains word-groups, but it must consist of words.

An overall description of every type of sentence member,word-group,clause and sentence is the purpose of practical grammar,where, as a rule,patterns of sentences and their members are treated on the basis of uni-versal notions valid for every language,and the description of syntacticphenomena characteristic of English is usually limited to the enumera-tion of different connectives.

a) The Sentence MembersSentence members are positional and relational syntactic units; they

mark the position of a word or word-group in the linear chain of a sen-tence (clause) in their reference to each other, their mutual relation onthe basis of most abstract five meaningful syntactic relations, such aspredicative (predicatival), attributive, adverbial, object(-ival) and sub-ject(-ival). E.g.: in the sentence He discovered that book on the uppershelf, he is the subject because it stands in subjectival relation to thepredicate verb discovered which stands to it in predicative relation. Sub-jectival and predicatival relations are reciprocal, bilateral. That stands inattributive relation to the object book (onesided, unilateral relation), bookis the object of the verb, to which it stands in objectival relation (alsounilateral), that book is a word-group as object, on the upper shelf isanother group as adverbial (unilateral relation).

A predicate does not necessarily consist of a verb only, but onlyverbs have objects (in English), that is why describing syntactic units, itis often necessary to use the morphological term verb.

75

Sentence members dependent on the main three (subject – verb–object) may form part of the latter when they are expressed by a groupcontaining a verbal form which may take an object or subject, e.g.: Wesat at table, the visitors waiting outside – secondary predication clause(group) as adverbial modifier.

As sentence members may be expressed by a word-group, the no-tions of subject-group, object-group, adverbial group, etc. are current insyntactic theory. This is an indication of close interrelation of sentencemembers and word-groups.

The idea of the reciprocity of subject – predicate relations may evokedoubts as to the status of the third main sentence element – the object.Maybe the verb – object relations are also bilateral, same as subject -predicate? The thing is that all verbs, especially in English, have a sub-ject, whereas only some subclasses of verbs have objects. The optionalcharacter of the object makes the relation of the verb to it unidirectional,the more so that the very linear structure of speech developing in timeshows the direction of dependence.

All the unilateral syntactic relations that bind together different mem-bers of the sentence are subordinating by nature. Co-ordinating relationsconnect homogeneous parts of the sentence: e.g.: The children and theparents went together. All subordinating relations are hierarchical.

The reciprocity of subject – verb (predicate) dependence may presenta controversial problem. Departing from the linear development of speechin time, the verb must be subordinate to its subject. In fact, it is the sub-ject that selects the form of the verb (he speaks), and not vice versa; butthe verb is the creator of predicative relations, the centre of predication.Predicative relations are directed towards the subject, the notion of pred-icativity presupposes a binary correlation of some action and its subject.Formally, the verb is subordinated to the subject but semantically thesubject forms part of a predicative nucleus, that is why it seems mostappropriate to consider subject – predicate relations as reciprocal therebeing no hierarchy of ranks within this dyad.

In contradistinction to word-groups, sentence members present amore abstract category; they reflect the logical meaningful relations withina sentence, the kind of interrelations of its constituents which make it acommunicative entity.

b) The Word-groupIf sentence members signal the position of constituents of a sen-

tence in their mutual interdependence, the constituents of a word-groupcombine according to the valency of words composing it. The co-occur-rence of words in one group depends on the valency of each word, i.e., itsinner potency to combine on the syntagmatic axis. Models (patterns) of

76

groups represent a certain system of combinations of words (and theirgrammemes) existing in a language. But even the patterns themselvesmore often than not require an indication of some semantic subclass oflexical units composing them, e.g.: of the type Vt + N1A + N2X which,filled by indicators of lexical subclasses means: verb transitive + nounanimate + noun inanimate = (in verbal representation) give him a book.

Among other syntactic units a word-group is the most concrete one;it reflects the facts characteristic of one particular language to a greaterextent than do sentence members and clauses. A sentence member and aclause, reflecting most abstract relations in speech are universal to a greaterextent than groups. Thus, groups must be described on the basis of factu-al material of a language with concrete lexical subclasses and sometimeseven separate words being constantly kept in view.

As is well known, the overwhelming majority of existing Englishgrammars, practical and otherwise, whatever principle they might be basedon, would give rules and descriptions of sentence members and sentenc-es (clauses), but not of word-groups. This fact finds explanation in thenature of groups. On the one hand, they are infinite and in the same wayas other syntactic units are associated with abstract patterns of word com-binations. On the other, it is difficult even to outline and enumerate thegeneral types of groups as a syntactic unit in the same way as it is donewith sentences and their members, the types and classes of which areexplicit in every grammar; e.g.: simple, complex, exclamatory sentenc-es, subordinate clauses, subject, object, etc. Groups of words bound to-gether by mutual combinability may cut across several sentence mem-bers, and also form a sentence. Especially it refers to groups formed bythe verb, the centre of predication.

c) The Clause (Sentence)The next syntactic unit on the rank scale is the clause or sentence, 'the

English language provides separate terms for the denomination of sentenceas an independent unit, and clause as a unit which possesses the structure ofa sentence with relations inherent to its constituents, being at the same timea dependent syntactic unit within a sentence. It is the sentence and not theclause that presents a communicative unit in human intercourse. The formalboundary of a sentence is the period (or other signs of syntactic pause). Allthe relations characteristic of syntactic units, such as predicative, attributive,etc., all members of the sentence, types of groups and clauses are reproducedwithin the limit of a sentence. In any syntactic unit larger than a sentence,such as a composite syntactic whole, a dialogue, a paragraph, etc. (suprasen-tential or supraphrasal units) the relations and units inherent in a sentenceare repeated, and the study of such units, outlimiting a sentence are conduct-ed on the basis of usual syntactic notions.

77

If attempts are made to disregard the types of syntactic units, theirhierarchy, their patterning in a sentence, and the syntactic relations amonglinguistic elements existing in a language as a communicative whole, thepicture of any syntactic unit becomes distorted. Such an erroneous repre-sentation of syntactic structures may result if, for instance, clipped sen-tences in dialogues or in any other supraphrasal unit are analysed as 'thingsinto themselves', on the level of words. E.g.: if in the unit: 'What time ishe coming?' – 'At five', at five is treated as a sentence expressed by anumeral,the nature of syntactic relation is obscured; in fact, however, atfive is an isolated adverbial modifier of time connected by adverbial re-lations with the verb of the first sentence.

The linguist undertaking a study on the supraphrasal level must bemainly concerned with the kind of correlations – cognitive, lexical, se-mantic, grammatical – that exist outside a sentence and among separatesentences, and with the concrete types of structures that prevail in a cer-tain supraphrasal entity.

d) Suprasegmental ElementsWhen reproduced in oral speech grammatical units are character-

ized by linguistic elements which are not reflected in their written repre-sentation, i.e. elements which cannot be segmented or isolated and di-rectly observed in the text. These are stress and intonational contour.They are suprasegmental elements and call for special study.

I.B.Khlebnicova. "Essencials of English Morphology" (p. 5 – 12)

78

3. MORPHEMIC STRUCTURE OF THE WORD

3.1. MORPHOLOGY AND MORPHEME

The word morphology itself consists of two meaningful elements:Gr. morphe – form, and logos – word, similarly these constituent ele-ments can be observed in some other units, like morpheme, allomorph.or biology. teology. archeology: another element -y recurs in history (Lat.historid), unity (O.Fr. unite), beauty (M.E. beaute = pretty). Although itsmeaning is rather abstract, it helps us to recognise the latter three wordsas nouns. The expression of plurality, f. ex., of objects, facts or humanbeings named by the nouns is achieved by the use of a special grammat-ical device – a morpheme in one of its realisations: [s] in patients' medi-cal histories: [iz] in unities of time, place and action; [z] in nouns arenames, etc. These number distinctions are regularly expressed in nounsto show the grammatical importance of the opposition of two forms of asingle category, number.

Morpheme is said to be the ultimate unit of the semantic level oflanguage. Morphemes are not divisible any further without breaking thewholeness of a word. Un-, under- in unusual, undergraduate are said tobe prefixal morphemes: -al, -ful in general, careful – are said to be thesuffixal morphemes, all these extending, changing or modifying the mean-ing of a root-morpheme: usual – unusual, careless – careful, undergrad-uate – graduate – postgraduate, etc.

In some cases the morphological analysis needs further historical,etymological inquiry, thus "wholeness" of admit when compared to ad-mit, admitted, admitting, becomes doubtful at the background of suchwords as commit, permit, dismiss, mission, missile. Prefixes ad-, com-,per-, dis- are historically determined, that is, became borrowed from Latintogether with prefixal Latin stems. Root-morphemes also admit varia-tion or sound change on morpheme boundaries in: admit – admission,but also in larger units, like was seeing, went down, have wept, etc. Allthese phenomena are the object of Morphology the latter being subdivid-ed into Morphology of Synthetic and Analytic Forms.

Morphology is that branch of linguistics which concerns itself withthe structure of words as dependent on the meaning of constituent mor-phemes and the system of morphological oppositions in a given languageincluding their grammatical categories as unities of form and content.For instance, the wordforms speaks and worked each consist of two mor-phemes: speak + s, work + ed. The left-hand parts of these words are

79

called lexical morphemes. They carry the lexical meaning of the wordsin question, whereas -s, -ed are grammatical mophemes, because theyserve to express the grammatical meanings of mood, tense, number, per-son and other grammatical morphological distinctions of the verb inModern English.

Morphology and morphophonology have that in common that a cer-tain unit acquires a meaning, becomes semiologically relevant, only inopposition with other units within the same system. (In contrast withwords, as units of Lexicology, where each one has got an individual ex-tralinguistic referent). With phonology, morphonology and morphologythe situation is much more complex: phonemes and grammatical mor-phemes have no individual extralinguistic referents, they become unitsof language only when mutually opposed, like [t] and [d] in dusk andtusk, or [-t] and [-d] in cried and asked.

3.2. SPECIAL NATURE OF THE MORPHEME

It is common knowledge that linguistics is essentially a quest formeaning.The units of feature-level, phonemes have no meaning of theirown, they only serve to differentiate the meanings of other units, theirfunction is confined to indicating "otherness", as in big vs (versus) pig,fig vs dig, let vs met and net, set vs shed, or did vs deed etc. Morphemesare the units of the semantic level, different types of morphemes fulfilldifferent functions being endowed with different types of meaning.

According to a dictionary 'meaning' is what is referred to or indicatedby sounds, words or signals. The concept of meaning in linguistics cannot be properly tackled unless we take into consideration the concepts ofplanes of expression and content of a linguistic sign.

Morphemes are linguistic signs of a very special nature. The studyof morphemes presupposes the study of their occurence, order,arrangement, combinability, mutual similarity or dissimilarity in asystemic way. The element -erne in morpheme points out to belonging toa system. Some morphemes are freer, some other less so, f. ex., grademay be found in a greater number of units, than under-, or -ate in termsof ordering: grade, gradient, degrade, degradation, undergraduate,gradually, while under before vowels: underestimate', before voiced andvoiceless consonanats: undermine, understand – but always precedingthe root, -less "takes" different types of roots but always following them:merciless, fruitless, speechless, colourless, etc.

The lexical morphemes may or may not directly correspond toobjects, facts, phenomena or properties of extralinguistic reality, their

80

meaning is more concrete or less so, but can be unmistakingly under-stood when looked at systemically. These units are semic and morphic,having their own individual meaning and admitting no morphologicalvariation: they are reproduced as it were in a number of characteristicpatterns, like in:

1) read-able 2) ir-relevant 3) ex-clus-ivethink-able ir-reverent ex-pens-iveeat-able ir-regular ex-haust-iveIt is with lexical morphology when the concept of partial phonetic-

semantic resemblance of morphemes stands out as a very important cri-terion. The words readable, thinkable, eatable are partially phoneticallysimilar because of the element -able reproduced in all three, as a resultthey become partially semantically similar because of naming a qualityof an object thought of or spoken about, in other words, they belong tothe same group of words – qualitative adjectives. The meaning of ir- inirregular, irrelevant, irreverent is understood as a result of opposing reg-ular and irregular, relevant and irrelevant, reverent and irreverent andthus revealing the positive (non-expressed) and the negative (expressed)implications. The fact that these three words belong to the same class ofadjectives is not dependent on the meaning of the prefixal element.

A grammatical morpheme has no partial-phonetic-semantic resem-blamce to any other form, being recurrent and intrinsically structured.

Grammatical morphology is sememic and allomorphic. It implies thatthe meaning of a grammatical morpheme of Number in nouns can be under-stood only through the narrow system of its realisations being positionallybound and determined; a number of positions presupposes a number of posi-tional variants, allomorphs, and their general meaning of Number, eithersingular or plural, in: dog – dogs, cat – cats [z], idea – ideas [zero]sing isopposed to [s] [z] brush – brushes [(i)z] impression – impressions, where[z] is chosen to be the main variant and morpheme representative being lessdependent on the quality of a preceding sound. As to the meaning of numberit remains very abstract and general in our understanding and comprehen-sion of multitude of cats, dogs, brushes and even of our own ideas and im-pressions. The meaning of plurality then is understood not individually butonly within a system of word-forms, paradigmatically.

There are words and their forms that are morphologically simple,like day, word. They may consist of a group of sounds or of one sound a,are. Forms like word, night, worth are morphologically primary, where-as, f ex., worthless, worthwhile, worthy, praiseworthy are derivative stemsand may produce lexical paradigms, i. e. revealing certain ways, models,patterns or examples of word-building. In derivative morphology theconcept of productivity is said to be of crucial importance.

81

Lexical derivative morphemes -less, -ness, -like are said to be high-ly productive, like in: timeless, countless, shameless; darkness, happi-ness, blondness, clumsiness, greyness, disinterestedness, insultingness,quickmindedness, etc; animal-like (behaviour), ball-like (structure); child-like, ladylike, shelllike, makelike, etc. Another group of derivative mor-phemes is described in linguistic literature as quasi-grammatical, beingalso highly productive: -able, -er, -ly,, f. ex., in capable, payable, fash-ionable, comfortable, changeable, perishable, etc; computer, philoso-pher, villager, sixth-former, three-wheeler, double-decker, happily, stu-pidly, widely, broadly, cowardly, scholarly, dayly, weekly, exceptionally,freely, rapidly, occasionally, regularly, etc.

Grammatical morphemes turn out to be almost absolutely produc-tive: -er, -est in adjectives like commoner, commonest, newer, newest; -ing in verbal forms, like coming, going, seeing, happening, doing, dis-covering, admonishing, believing, hoping, lying, soothing, activating,calculating, predicting, etc.

In Grammatical morphology what we are dealing with is form-build-ing and not word-building. Thus, to conclude: a morpheme is

1) a recurrent meaningful form which cannot be further analysedinto smaller recurrent meaningful forms;

2) a grammatical morpheme is a linguistic form which bears nopartial phonetic-semantic resemblance to any other form;

3) a morpheme is the smallest unit of the expression plane whichcan be correlated directly with any part of the content system: a mor-pheme is a group of two or more allomorphs which conform to certainusually rather clearly definable criteria of distribution and meaning. Theabsence of any special grammatical morphological expression (zero-morpheme) may also be meaningful if it serves to denote a relation be-tween the words (the word-forms) in asentence;

4) a morpheme is syntactically or positionally bound, it cannot takeand arbitrary position;

5) a morpheme is a unilateral unit, it never expresses both a lexicaland a grammatical meaning, while the lexical meaning is concrete, mate-rial, the grammatical meaning is general and abstract.

The expression of grammatical meaning is subservient to the lexi-cal meaning, it is additional to it. Grammatical meaning is recurrent andsystemic (forming part of asystem), the lexical meaning is free, indepen-dent, optional and individualised. It does not have to lean on anything.When saying "Morphemes are ..." we are free to begin by saying thisword or "elements", "segments", "units", etc., but we cannot avoid spec-ifying it as "one" or "more than one";

82

6) a morpheme is of historical nature, and what is or was singledout as a morpheme now or yesterday, may cease to be so tomorrow, f. ex.-ous in glamourous does not stand out so obviously in tremendous orstupendous, the relationship between form and meaning, content andexpression in linguistic units does not lend itself to a neat compartmen-talisation because language remains in a state of constant change lessvisible in Grammar, however, than in Lexis.

The morphological structure of words and the morphological systemof language as a whole becomes as time goes on, affected by a great varietyof extralinguistic factors, military invasions, movements for independence,economic growth, level of education, roots of literacy etc., etc.

3.3. WORD AS A NOMINATIVE UNIT

The word is a basic nominative unit. Without words there cannot beany communication even in thought, to say nothing about speech com-munication.

From the point of view of its nominative function, the word is anelementary indivisible constituent part of the lexicon.

It is not easy to identify the word because the words are heteroge-neous from the point of view of both content and form.

To find the criteria of word identification linguists resort to the no-tions of functional correlation and continuum. Functional correlation con-nects the elements which have similar and different properties. In fact,within a complex system of interrelated elements there exist two types ofphenomena – "polar" and "intermediary". Polar phenomena stand to oneanother in an explicit opposition. Intermediary phenomena are located inthe system in between the polar phenomena, making up a gradation oftransitions. A total of these transitions makes up a continuum. Thus, be-tween proper nouns and common nouns – polar phenomena – there existdifferent transitions of semi-proper nouns which make up a continuum.

Giving a definition to the word on these lines, it is necessary todescribe the notional one-stem word and the grammatical morpheme asthe opposing polar phenomena. The continuum existing between them isconstituted by functional words. Functional words are very limited innumber and perform various grammatical functions. In distinction to these,notional words are infinite in number and are nominative units proper.Thus, the word is the nominative unit of language built up by morphemesand indivisible into smaller segments as regards its nominative function.

M.Y.Bloch "Theoretical English Grammar" (p. 65 – 67)

83

3.4. MORPHEMIC STRUCTURE OF THE WORD

The morphological system of language reveals its properties throughthe morphemic structure of words. So, it is but natural that one of the essen-tial tasks of morphology is to study the morphemic structure of the word.

In traditional grammar the study of the morphemic structure of theword is based upon two criteria – positional and semantic (functional).The positional criterion presupposes the analysis of the location of themarginal morphemes in relation to the central ones. The semantic criteri-on involves the study of the correlative contribution of the morpheme tothe general meaning of the word. In accord with the traditional classifi-cation, morphemes at the upper level are divided into root morphemesand affixal morphemes (lexical and grammatical).

The morphemic composition of modern English words has a wide rangeof varieties but the preferable morphemic model of the common Englishword is the following: prefix + root + lexical suffix + grammatical suffix.

Further insights into the correlation between the formal and func-tional aspects of morphemes may be gained in the light of the "alioemic"theory put forward by Descriptive Linguistics. In accord with this theo-ry, lingual units are described by means of two types of terms – "allo-terms" and "eme-terms". Eme-terms denote the generalized, invariantunits of language characterized by a certain functional status, e.g., pho-nemes, morphemes, lexemes, phrasemes, etc., but in practical analysisthis terminology is applied only to the analysis of phonemes and mor-phemes. Allo-terms denote the concrete manifestations or variants of theeme-units. Allo-units are distinguished by their regular co-location withother elements of language. Typical examples of allo-units are allophonesand allomorphs.

The allo-emic identification of lingual elements forms the basis for theso-called "distributional" analysis. The aim of the distributional analysis isto study the units of language in relation to the adjoining elements in the text.

In the distributional analysis three main types of distribution arediscriminated: contrastive distribution, non-contrastive distribution, andcomplementary distribution. Contrastive and non-contrastive distributionsconcern identical environments of different morphs. The morphs are saidto be in contrastive distribution if their meanings (functions) are differ-ent; such morphs constitute different morphemes, e.g., "returned // re-turning // returns". The morphs are in non-contrastive distribution if theirmeanings (functions) are identical; such morphs constitute "free alter-nants" ("free variants") of the same morpheme, e.g., the suffixes "-ed"and "-t" in the verb forms "learned // learnt", or the suffixes "-s" and "-i"in the noun forms "genies // genii". As for complementary distribution, it

84

concerns different environments of formally different morphs which ful-fil one and the same function; such morphs are termed "allo-morphs",e.g., there exist a few allomorphs of the plural suffix of the noun: "-en"(children), "-s" (toys), "-a" (data), "-es" (crises), "-i" (genii), the zeroallomorph (trout // trout), etc.

The application of distributional analysis to the morphemic levelresults in the classification of morphemes on distributional lines. In ac-cord with this classification a few "distributional morpheme types areidentified: free and bound morphemes, overt and covert morphemes, ad-ditive and replacive morphemes, continuous and discontinuous mor-phemes, segmental and supra-segmental morphemes.

M.Y.Bloch "Theoretical English Grammar" (p. 67 – 68)

3.5. MORPHOLOGICAL PROCESSES

3.5.1. Processes affecting grammatical formThe units of morphological level do not function in isolation in a flow

of speech, since they – morphemes as well as word-forms – are only theconstituent elements, the former for the words, the latter for the sentences,both types forming a kind of hierarchy of subservient structural elements.

When carrying out a grammatical-morphological analysis twoconcepts were found to be of great help and worth noting. Juncture is awayof joining elements, amalgamating them into one global whole, whilediarheme is away of cutting them, breaking them, keeping them apart, thesetwo terms are most fruitfully exploited in Syntax. For the purposes ofmorphological analyses two main processes play a considerable role ininflexional morphology, or morphology of synthetic word-forms.

Within grammatical inflexional morphology what we are interestedin are the processes that characterise morpheme-structure proper andmorpheme boundaries. Morphonology of these units concentrates ondiscovering and formulation the rules for the transformation ofphonological sequences into morphological ones. The phonological non-identity of the different sound (and orthographic) complexes signals theirfunctional morphological identity. It applies the analysis of the innerstructure of the grammatical expression of the plural number in nouns,3rd person singlular of Present Tense forms in verbs, Past Tense forms inverbs. It is well known that if the stem ends in a strong (fortis) consonant,the inflexional morpheme is always [-s]; in the case of the weak (lenis)consonant it is [-z]. After sibilants the same moipheme has an altogether

85

different sound envelope [-iz], where -/- element is regarded as an interfix,a connecting element, which belongs neither to the stem, nor to theinflexion, and is regularly inserted when the phonetic realisation of agiven sequence is antropophonically impossible.

On morpheme boudaries not all the distinctive features of the soundare realised, some sound properties may become weakened, neutralised,some other strengthened, thus leading to fusion. In our case the archiphoneme <z> which functions as the grammatical morpheme of 3rd person,singular in verbs is a bundle of distinctive features of fricative, groove,alveolar. Functionally speaking the semiological function of the positionalvariants of the corresponding morpheme is based on a single distinctivefeature – the opposition of strong versus weak, as in, f. ex.:

A. Her taste in music coincides with her husband's. If you want togo by bus it suits me fine.

He digs all his information out of books and reports. She takes herchildren to school by car.

B. One of the gang blabbed to the police and they were all arrested.He swiped at the ball and missed.

I really sweated over this essay.Fusion can as well be seen on word-form boundaries in the flow of

speech, as in When I was eleven I was sent to the secondary school. I'mmissing you so much,you know?, etc. Agglutination does not presupposeany change in the quality and the quantity of the neighbouring soundswithin a word-form or on the boundaries of the morphemes, like in come+ ing, cry + ing, stick + ing, miss + ing, etc.or in rapid + ly, scarce + ly,main + ly, etc.

The formation of the plural in nouns is accompanied by thealternation of a final consonant in calf – calves, elf – elves, half – halves,thief – thieves, turf – turves, wife – wives, wolf – wolves. lt can be as wellaccompanied by vowel-change as in: foot – feet, goose – geese, woman –women, louse – lice, mouse – mice, tooth – teeth. Another type ofgrammatically pertinent variation is observed in the following word-pairs:

sane sanity divine divinityvain vanity sublime sublimityurbane urbanity sentile sentilitymendacious mendacity finite infinitystate static lyre lyricangel angelic parasite parasiticvolcano volcanic

86

verbous verbosity serene serenityatrocious atrocity meter metriccone conic athlete athleticepisode episodicPlato Platonic

Although in all these cases we deal with the lexical units (and lexi-cal morphenes) the observed regularities are not of lexical nature, theystep out from the domain of lexis and approach Grammar; functionallythey help to differentiate the lexical-grammatical properties of the wholeclasses of words – nouns and adjectives, thus, the meaning of those chang-es is of a more abstract general character. These cases may be regardedas borderline ones between compounding (snow-white, catlike) and in-flexion proper (hopes, ideas, stepped out, regarded, etc.).

O.V.Alexandrova, T.A.Komova "Modern English Grammar:Morphology and Syntax" (p.16 – 18)

3.5.2. Processes affectinggrammatical content

The relationship between linguistic form and linguistic meaning re-veals itself in what is known as syncretism and suspension, or in otherwords, grammatical homonymy and grammatical synonymy. The latter issometimes terminologically presented by "variation of expression", "re-dundance of expression plane", "multiplication of distinctions", "emptydistinctions on the expression plane with no correlation of the plane ofcontent", etc. Despite the obvious cases of syncretism and suspension thegeneral rule remains valid. Normally identity (sameness) on the expres-sion plane points to the identity on the plane of content, and difference onthe expression plane signals difference (otherness) on the plane of content.

By syncretism is meant falling together of two or more grammaticalmeanings in one and the same grammatical (inflexional) form. This factbecomes manifest mainly as dependent on the vigour of the non-syncre-tisized inflexional forms, their ability to induce or to evoke in the formerthe particular content.

In Russian nouns, f. ex., the forms of the Nominative and the Accusa-tive in the declension-type of мышь are homonymous, they coincide intheir expression, but remain different in their content. This statement canonly be made because in Russian a paradigm of a noun normally has sixinflexional forms: -а, -ы, -e, -у, -ой, -e, hence the syncretised forms are

87

discernible against the background of the non-syncretised ones. This, how-ever, can not be said about the English noun-form man in the function ofthe subject or the object. This form is not a result of any syncretism, be-cause there are only two case forms in Modern English. Morphology ofsynthetic forms is, in general, based on inflexional homonymy and zero-morphemes make up for paucity of inflexional forms.Thus, in verbs:

1. He knew that it's there somewhere.2. If he knew that all before.3. I was a teacher when I was 18.4. If I were / was a teacher I would not complain.forms of the Past Indefinite and the Subjunctive: knew, knew, was and

was are homonymous; while was and were are symonymous, they are, al-though different in form, identical in the grammatical content they render.

Another example. The morpheme [z] can function as 1) the plural ofnouns in: dogs, cats, clashes; 2) the possessive case inflexion in nouns:dog's barking, a dog's life, cat's-eye, the cat's whiskers; 3) the plural forminflexion in demonstrative pronominal adjectives this – these, that – those,where the interchange of [s]/[z] is supported by the interchanging vowel inthe root morpheme. Within the grammatical system of the English verb wealso have the alternation of a sound of the root morpheme with a zero-morpheme, like in (I) think – (I) thought, 4) the absolute form of pronounslike hers, ours; 5) the verbal inflexion of the 3 person singular, presenttense, as in: (he) speaks, declares, proposes. The same sound (or sounds)in grammatical morphology obeys the rules of homonymy at large. Wespeak of homonymy when the same element of sound, the same unit of theexpression level is connected with different units on the semantic level.Homonymy presupposes that the grammatical meaning of two forms in-compatible, while the form is assumed to be identical. The grammaticalmeanings of the possessive in nouns and 3rd person singular in verbs areincompatible, they cannot be brought together as variants of the same form.Thus, in [z1] as the plural of nouns, [z2] as the possessive case, [z3] a; theverbal inflexion, all three grammatical inflexions are identical in form butoperate within a corresponding paradigmatic set.

Synonymy as a fact of grammatical morphology presupposes thattwo units have the same grammatical meaning being different in form,and the above case of was /were in If I was/were a teacher I would notcomplain was given as an illustration.

Grammatical polysemy is observed in grammatical expression of a hostof intricate distinctions of a noun known as the genitive or the possessive.

Things, objects, events, human beings can be specified as belong-ing to, or associated with, or connected with, as in, for example:

88

A. John 's motorbikeher grandmother's houseB. the dog's head John's armC. his friend's reactionMrs Thatcher's greatest errorD. the car's colour and designE. the country's biggest city the city's populationIn some cases the possessive form of anoun functions in a similar

way to apossessive pronoun:Her hand felt different from David's.Her tone was more friendly than Stryke's.It is your responsibility rather than your friendsSometimes the idea of possessivity becomes rather abstract:women's magazines (magazines for women to read) the men's lava-

tory (to be used by men)a policeman's uniform (that makes them different from soldiers or

navy officers, etc.)The same grammatical inflexion may refer to a someone's home or

place of work:He's round at David's.She stopped off at the butcher's for a piece of steak.Sometimes ownship is specially emphasized lexically:Cf We must depend on his own assessment.We must depend on David's own assessment, etc.Thus all the above cases are brought together as semantic variants

of the same grammatical meaning of possessivity, they are considerednot to be incompatible, but polysemous, rendering the ideas of posses-sivity proper, partitivity, association, connection, qualification, location.Some of these forms become semantically close to polysemantic prepo-sitional stuctures with of, for, in, that of, belonging to.

O.V.Alexandrova, T.A.Komova " Modern English Grammar:Morphology and Syntax " (p. 18 – 20)

3.6. STRUCTURAL FEATURES OF PRESENT – DAY ENGLISH

In the course of its development, English, as compared to Anglo-Saxon, its parent language, has changed beyond recognition: it has lost,with the exception of a few remains, most of its inflexions. What used tobe expressed by inflected noun forms is now expressed either by meansof prepositions, i.e. lexically, or by a special position in the sentence orthe word-combination, i.e. syntactically.

89

The greatest changes of all can be seen in the adjective and theverb. The elaborate declension system of the adjective has completelydisintegrated, and the adjective is now an invariable part of speech (notmentioning degrees of comparison). To quote B.Ilyish [12, p. 307], "thesimplification of adjective morphology had to be ‘paid for’ by limitationof freedom in word order". The verb has undergone radical changes aswell: personal inflexions, with the exception of the singular 3rd personpresent tense inflexion, have been lost; new tense forms have come intobeing: present-day English now boasts of 16 tense forms against two inAnglo-Saxon. To innovations we should attribute the passive forms, theanalytic forms of the subjunctive mood, and others. All these modificationshave changed the structure of English: present-day English is generallydescribed as an analytic language. This statement is not precise, for Englishis still in the process of development. We can still observe the strugglebetween the old and the new, i.e. between synthetic and analytic forms.

At this stage, English is a predominantly analytic language. Thestruggle between the old and the new can be seen in many areas of English.Consider, for instance, the formation of feminine nouns. Since thebeginning of the 13th century, together with the decay of grammaticalgender, English has gradually lost the unrestricted power of formingfeminines by inflexions and has replaced the morphological process bythe syntactic, or analytic process, i.e. by the addition of words denotingsex to the noun:

A visitor – a gentleman visitorA servant – a woman servantAn employee – a female employeeA teacher – a male teacherA cat – a she (lady) cat; a Tom-catThe process of the replacement of old synthetic forms is also seen

in the use of the forms who/whom. The inflected, or old, form whom isdisappearing from the spoken language and being replaced by who. It isquite normal to say I don’t know who to invite, though in the writtenlanguage we still find whom.

Whom seems to be unshakable in one position, viz. after apreposition. Cf.

To whom shall I give it? vs.To who shall I give it?I don‘t know for whom it is intended. vs.I don’t know for who it is intended.In the spoken language, sentences with whom are not usually used;

they occur with who and the prepositions to, for, with in sentence finalposition:

90

Who shall I give it to?I don’t know who it is intended for.Analytic processes are also seen in the formation of the comparative

and the superlative of adjectives, where forms with -er and -est are beingreplaced by forms with more and most, e.g. commoner – more common,commonest – most common. Other adjectives with more and most includecloudy, fussy, cruel, quiet, subtle, clever, profound, simple, and pleasant.The spread of more and most can be illustrated by forms like more well-informed and most well informed or more well-dressed and most well-dressed, where people used to say better-informed/ best-informed and bet-ter-dressed/ best-dressed. More and most are spreading even to monosyl-labic adjectives: crude – more crude, most crude; plain – more plain, mostplain; keener – more keen, most keen.

Analytic processes are also going on in the ‘realm’ of the verb. Forexample, the distinction formerly made between shall and will is beinglost, and will is coming to be used instead of shall when the meaningintended is simply futurity (i.e. prediction):

I will be there.We will do it.The same can be said about should and would:I would like to know.We would like to know.However, there are some co-texts where should cannot be replaced

by would:I insist that he should come with us.I asked the man whether the boy should wait.Should is being replaced by would in purpose clauses:I lent him the book so that he should/would study the subject.New auxiliaries are coming into existence, for instance, get and want.

Get is used in forming a passive:He got hurt. vs. He was hurt.Suppose someone gets killed. vs. Suppose someone is killed.He failed to get re-elected. vs. He failed to be re-elected.Want is used in the spoken language to mean ought to, must, or

need:You want to be careful what you‘re doing.You want to go to a doctor.You want to take it easy.Another analytic process that is going on in English concerns the

use of the verb have. When have is a full verb (i.e. when it means posses-sion), not an auxiliary, it forms its negative and interrogative construc-tions in two ways: 1) with the auxiliary do (e.g. Do you have a car? He

91

did not have a car.) and 2) without the auxiliary do (e.g. Have you a car?He hadn’t a car.) Under the circumstances, British English would oftenadd got: Have you got a car? He hadn’t got a car. The use of the auxilia-ry do in possessive constructions is due to American usage. When havemeans experience, only the analytic construction is used – in both Ameri-can and British English: I had difficulty breathing.

Special mention should be made of the analytic construction con-sisting of verb + verbal noun: to have a look at, to give a ring, to make amistake. This is a construction that is gaining ground rather fast in present-day English. The usefulness of the constructions becomes obvious whenwe compare them with the corresponding synthetic constructions:

A.He looked at his watch. vs. He had a look at his watch.He walked in the park. vs. He had a walk in the park.He danced at a discotheque. vs. He had a dance at a discotheque.B.He had a bad dream. vs. He dreamed badly.He made a terrible mistake. vs. He mistook terribly.The comparison of the constructions shows that they ‘come in handy’

when we have to express the completion of the process and when wewish to modify the process in ways other than the synthetic form can bemodified. In other words, the nominalized verb has more expressive pos-sibilities than the corresponding verb.

The struggle between the old and the new is a perpetual process.The old forms will not give in so easily. Some of them, for instance, theold genitive seems to be strongly entrenched and even to be regaininglost territories.

According to Charles Barber (1964), the old genitive has come intocommon use with nouns denoting inanimate entities. The scholar illus-trates the statement by such constructions as biography’s charm, therecord’s imperfection, evil’s power, criticism’s standard, human nature’sdiversity, amendments to the game’s laws. The reason for this must lie inthe simplicity and brevity of such constructions as compared to the cor-responding analytic ones.

The inflexions that remain in English (the third person singularpresent tense inflexion of the verb, the past tense inflexion, the past par-ticiple inflexion, the -ing inflexion, the plural and genitive inflexions ofthe noun, the nominative – accusative contrast in the personal pronouns(we/us; he/him), to quote Barber, "show little sign of erosion". The re-maining synthetic forms, however, cannot overshadow the general pic-ture: English has come to rely more and more on function words (prepo-sitions, auxiliary verbs) and word-order in expressing meaning.

92

3.6.1. The analytic and synthetic formsA form is analytic if the grammatical meaning is realized by a gram-

matical word-morpheme, e.g. John has done his work, where the gram-matical meaning of the form has done is expressed by the grammaticalword-morpheme has. In a broad treatment of analytic constructions, thegrammatical meaning may be expressed by words which are not devoidof their lexical meaning, e.g. the man of property or we work, where thegrammatical meanings of subordination (or modification) and person arerealized lexically by the preposition of and the pronoun we, respectively.

To grammatical word-morphemes we can assign the following: have(John has done the work or John had a dream), be (The work is beingfinished), do (He does not love her). The status of shall and will is con-troversial: there is a view that shall and will have a lexical meaning. Thestatus of such constructions as more beautiful and most beautiful is con-troversial, too: it is not clear whether more and most are adverbs, i.e.notional words, or grammatical word morphemes.

Present-day English possesses four synthetic forms, i.e. forms which arean inseparable part of the word. There are three types of such forms in English:

1) outer forms, or outer inflexions, i.e. book + s – books; clever +er – cleverer; clever + est – cleverest; walk + s – walks; walk + ed –walked; walk + ing – walking.

2) inner forms, or inner inflexions (sound alternation), e.g. mouse –mice; man – men; write – wrote – written; sing – sang – sung.

Sound alternations do not play a significant role in Modern En-glish; their role has been greatly reduced as compared to Old English.

3) suppletive forms, e.g. good – better – best; bad – worse – worst;go – went; I – me, mine; we – us, our.

In a broad treatment, suppletion – a morphological process in whichone form wholly replaces another – is of two types: full and partial. Forexample, went illustrates full suppletion (go vs. went) while thought illus-trates partial suppletion (think vs. thought). Go and went, think and thoughtare considered as two forms of the same words because in the vast majorityof verbs the past tense is derived from the same stem as the present or infin-itive. "It is against this background that the units go and went come to beconsidered as forms of one word, formed from different stems" [12, p. 26].

In Modern English suppletive forms constitute a very insignificantelement. Yet they present very widely used words. It will be obvious thatthese synthetic forms are not productive in present-day English. This isespecially true of inner and suppletive forms. The low productivity of theforms is compensated for by their universality: one and the same grammat-ical element is often employed to build different grammatical forms.

93

3.6.2. Analytic FormsThe grammatical system of Modern English is mainly based on what

is usually described as analytic forms, that is, combinations of the typemore rapidly, most tiresome, has declared, had acknowledged, was in-terviewed, have been invited, etc.

In the orthographic version an analytic form appears as a combina-tion of two or more elements, written separately; in the flow of speechthey are articulated as one global whole and their disintegration is al-ways specially prosodically-syntactically grounded. The function of ananalytic form is equivalent to that of a single synthetic word-form.

The existence of a series of synthetic forms of a word is a necessaryprerequisite for any discussion of analytic forms in this or that subsystemof English. The fact that there are synthetic forms of comparison in adjec-tives is crucial for interpreting the combinations of more/most + adj, as itsgrammatical word-forms, like in tall – taller – tallest, soft – softer – soft-est, heavy – heavier – heaviest; handsome – more handsome – most hand-some, obliging – more obliging – most obliging, unfortunate – more unfor-tunate – most unfortunate. The fact that a similar analogy does not exist inEnglish for expressing a less or diminishing quality or property as namedby an adjective remains decisive in our treatment of less/least + adj struc-ture as a free word combination.

In verbs a situation turned out to be very complicated due to the factthat analysity has been taking an upper hand during the course of itsdevelopment as a system. For a combination of words to become an ana-lytic form, that is, apart of grammatical morphology, several essentialconditions must be fulfilled. An analytic form is a word-combination ofa specific character. lt does not consist of two full lexical units but of twowords, one of which is syntactic, syncategorematic word and the other –the main, categorematic word which carries the burden of the full lexicalmeaning of the construction. In other words, neither lexical nor gram-matical meaning of an analytic form can be attributed to one of the ele-ments of its structure, both belong to this structure as a whole.

It must be mentioned that the meaning of the two parts should alsobe compared on the basis of greater or less abstraction. The lexical mean-ing of written in have written is concrete, whereas the meaning of haveis, comparatively speaking, abstract. It can be defined only in terms ofthe opposition between have written and wrote, have written and hadwritten. Moreover, the combination as a whole cannot be explained onthe basis of inividual extralinguistic reference, because have written doesnot mean I have something written (and not just outlined). Otherwisestated, the composite nature of have written on the expression plane and

94

its function on the semantic level make it indistinguishable from that ofgrammatically specified synthetic form proper.

In reality of speech there do exist cases when analytic forms tend tobecome less integral, less global, freer and more like a word combination.

A grammatical form is said to become isolated when a grammaticalform does not exist in a combination with any other form than that underconsideration, f. ex. been in a combination with has/have/had been. lt isunknown outside this type of structure.

A grammatical form is said to be isolated if its grammatical meaningbecomes different outside a combination in question, for example pan-elled in have panelled has a different grammatical loading than panelle inhe should have his drawing-room panelled (a) and not painted or decorat-ed with wall-paper, b) or he bought his drawing-room panelled, etc.).

Sometimes, however, the combinabilitv of alexically loaded ele-ment with a grammatically loaded one cannot be explained, and theystand out as "isolated". F. ex., there is a combination of two non-finiteverb forms: infinitive + participle I, but not participle I + participle I likein to be going and not being going.

Combinabilitv of a lexically loaded element and a grammatical load-ed element (of the same word) is possible, but the series of these formsstand apart in their general meaning in the same syntactic environment,f. ex., both forms of will are found in combinations like will do and woulddo, but the former functions within the Indicative Mood, while the latterwithin the Oblique Mood System. For example: By next year all the moneywill have been spent. (Indicative). If I had seen the advertisment in timeI would have applied for the job. (Irreal Conditional).

In some other cases it is possible to speak of homonymuos twinstructures, where a lexical element may or may not become grammati-cally morphologically isolated, thus in a combination this sound is pala-talised we can speak of an analogy between this pattern and patternswith adjectives proper – the sound is/becomes palatal/voiceless, the soundremains unreduced, etc. The use of the adverbial modifier transfers thewhole structure from the nominative into the verbal one, f. ex., in: Thepast tense morpheme in a position after a weak plosive <d> is constant-ly voiced and frequently palatalised. This type of isolation of a word-combination is lexically-grammatically bound.

To sum-up: within grammatical morphology there are no analytic word-forms of the words. These forms may be analytic and synthetic along dif-ferent lines of linguistic interpretation. Thus, in a form f Present or PastPassive, the idea of time is expressed synthetically, through one singleelement, while the ideas of Passivity (or Voice) is shown analytically, bymeans of both the auxiliary verb and the Suffix – of the Participle II.

95

Originally analytic forms are all free combinations of words, whichhave gradually become isolated under the stress of the morphologicalsystem of English.

It is the dual nature of analytic forms which explains the existenceof a considerable number of borderline cases. On the one hand, certaingrammatical contexts make for the vagueness of the borderline betweenanalytic forms and word-combinations. On the other, there are some com-binations in the English language (such as shouldn't + infinitive) whichfigure as analytic forms in some cases, while in others – tend to be indis-tinguishable from ordinary modal word-combinations.

This point deals with the relationship between an analytic form anda phraseological unit. When we speak of a certain unit as phraseological,we assume that it is characteriesed by lexical semantic isolation, by idi-omaticitv. The global meaning of a phraseological unit is not deduciblefrom the simple sum of the meanings of its parts.

Although analytic forms are also global units, they differ from phra-seological units in the sense that, in contrast with the latter, the speakerdoes not have to know beforehand what their global particular idiomaticmeaning may be. No previous information is necessary for forms like tobe annihilated, backtracked, was demoted, has depleted to be fully gram-matically comprehended. In other words, if phraseological units are anal-ogous to words as lexical units, analytic forms are equivalents of thegrammatical forms of words.

O.V.Alexandrova, T.A.Komova "Modern English Grammar:Morphology and Syntax" (p. 20 – 22)

96

4. CATEGORIZATION IN MORPHOLOGY

4.1. NOTION OF OPPOSITION.OPPOSITIONS IN MORPHOLOGY

The most general meanings rendered by language and expressed bysystemic correlations of word-forms are interpreted in linguistics as cat-egorial grammatical meanings. The forms rendering these meanings areidentified within definite paradigmatic series. The grammatical categoryis a system of expressing a generalized grammatical meaning by meansof paradigmatic correlation of grammatical forms. The ordered set ofgrammatical forms expressing a categorial function constitutes a para-digm. The paradigmatic correllations of grammatical forms in a categoryare exposed by grammatical oppositions which are generalized correla-tions of lingual forms by means of which certain functions are expressed.

There exist three main types of qualitatively different oppositions:"privative", "gradual", "equipollent". By the number of members con-trasted, oppositions are divided into binary and more than binary. Theprivative binary opposition is formed by a contrastive pair of members inwhich one member is characterized by the presence of a certain featurecalled the "mark", while the other member is characterized by the ab-sence of this differential feature. The gradual opposition is formed by thedegree of the presentation of one and the same feature of the oppositionmembers. The equipollent opposition is formed by a contrastive group ofmembers which are distinguished not by the presence or absence of acertain feature, but by a contrastive pair or group in which the membersare distinguished by different positive (differential) features.

The most important type of opposition in morphology is the binaryprivative opposition. The privative morphological opposition is basedon a morphological differential feature which is present in its strong(marked) member and is absent in its weak (unmarked) member. Thisfeaturing serves as the immediate means of expressing a grammaticalmeaning, e.g. we distinguish the verbal present and past tenses with thehelp of the privative opposition whose differential feature is the dentalsuffix "-(e)d": "work // worked": "non-past (-) // past (+)".

Gradual oppositions in morphology are not generally recognized;they can be identified as a minor type at the semantic level only, e.g. thecategory of comparison is expressed through the gradual morphologicalopposition: "clean//cleaner//cleanest". Equipollent oppositions in Englishmorphology constitute a minor type and are mostly confined to formal

97

relations. In context of a broader morphological interpretation one cansay that the basis of morphological equipollent oppositions is suppletiv-ity, i.e. the expression of the grammatical meaning by means of differentroots united in one and the same paradigm, e.g. the correlation of thecase forms of personal pronouns (she // her, he // him), the tense forms olthe irregular verbs go // went, etc.

As morphological gradual and equipollent oppositions can be reducedto privative oppositions, a word-form can be characterized by a bundle ofdifferential features (strong features) exposing its categorial properties.

M.Y.Bloch "Theoretical English Grammar" (p. 68 – 69)

4.2. OPPOSITIONAL REDUCTION

Oppositional reduction, or oppositional substitution, is the usage ofone member of an opposition in the position of the countermember. Fromthe functional point of view there exist two types of oppositional reduc-tion: neutralization of the categorial opposition and its transposition.

In case of neutralization one member of the opposition becomesfully identified with its counterpart. As the position of neutralization isusually filled in by the weak member of the opposition due to its moregeneral semantics, this kind of oppositional reduction is stylistically co-lourless, e.g.: "Man is sinful." It is an example of neutralization of theopposition in the category of number because in the sentence the noun"man" used in the singular (the weak member of the opposition) fulfilsthe function of the plural counterpart (the strong member of the opposi-tion), for it denotes the class of referents as a whole.

Transposition takes place when one member of the opposition placedin the contextual conditions uncommon for it begins to simultaneouslyfulfil two functions - its own and the function of its countepart. As aresult, transposition is always accompanied by different stylistic effects,e.g.: "Jake had that same desperate look his father had, and he was al-ways getting sore at himself and wanting other people to be happy. Jakewas always asking him to smile" (W. Saroyan).

In the cited example the transponized character of the continuousform of the verb is revealed in its fulfilling two functions – one of themis primary, the other is secondary; the primary function of the said verbform is to denote a habitual action, while its secondary function consistsin denoting an action presented in the process of development. Due tothe transpositional use of the aspect verbal form, the analyzed contextbecomes stylistically marked.

98

The study of the oppositional reduction has shown that it is effectedby means of a very complex and subtle lingual mechanism which evolvesthe inherent properties of lexemes, lexical and grammatical distributionof the replaced word-form and numerous situational factors such as theaim of communication, the speaker's wish either identify or to character-ize the denoted object, to reveal some facts or to conceal them, to soundeither flat or expressive, the speaker's intention to evaluate the discussedobjects, the interlocutors' sharing or nonsharing of the needed informa-tion, etc. All these factors turn oppositional reduction into a very power-ful means of text stylization.

M.Y.Bloch "Theoretical English Grammar" (p. 68 – 69)

4.3. CATEGORIZATION: CATEGORIES,CATEGORIAL FORMS

Outside Grammatical morphology category is defined as operatingin a system for dividing things according to appearances, qualities, etc(Cambridge International Dictionary of English, 1995); a type, or a groupof things having some features that are the same; Category, a class, or agroup of people or things regarded as having certain features, etc. incommon: (Oxford Advanced Learner's Dictionary, 1995, p. 176). Accord-ingly to categorise means to place sb/sth in a category (ibid, p. 176).Thus, categorization turns out to be very much like labelling, that is de-scribing or classifying sb/sth (ibid, p. 658).

In the New Webster Encyclopedic Dictionary of the English Lan-guage the word category is defined as follows: Category, n, (Greek katego-ria, a class or category, from kategoreo – 'to accuse, to show, to demon-strate'; -kata, 'down' and agoreo 'to speak in an assembly, from agora, aforum, or market' . One of the highest classes to which objects of thoughtcan be referred; one of the most general heads under which everything thatcan be asserted of any subject may be arranged; in a popular sense anyclass or order in which certain things are embraced [1971, p. 128].

This definition is close to what can be found in philosophy: a cate-gory is a reverberation in the human mind of the most general propertiesof objects. Categories are derived from observation of the objects andtheir properties, they are rational abstractions and generalisations of thehuman mind.

Thus, categorization is the process of discovering and explainingthe fundamental concepts of a science, the first step in the elaboration ofa general scientific theory. Categorization is closely related to a method-

99

ology of a science, the way of interpreting scientific facts, including lan-guage, its ontology and epistemology.

In philosophy the categories of matter, time, space, motion, nega-tion, etc. are the most general concepts that are the reverberations ofphenomena of objective world in our consciousness, they are the resultand the steps in the process of our cognition of the world around us.These categories are of universal character and reflect properties andrelationships proper to all natural and social phenomena.

In every concrete field of human knowledge (language as well) cat-egories present the result of a very specific kind of reverberation of theparticular aspects of objective reality. Within the domain of languagecategories are the broadest and most general characteristics of linguisticphenomena, but not necessarily permanent and stable; language is in astate of flux and its dynamic character is also reflected in the generalisa-tions made on the categorial level.

The -emic character of language is not superimposed by our analyt-ic thinking, the system of levels and the system of categories reflect theproperties and the relationships which the language has acquired at aparticular stage of its development. For instance, the category of semio-logical relevance may be considered to characterise the feature level ofthe English language during the whole period of its development, whilethe category of Taxis (or simultaneity – anteriority) reflects the proper-ties of the English verb at a later stage in its history.

It's worth repeating that language is primarily a means of communica-tion, the most important aim of it consisting in the passing on of information,in exchanging the "shared meaning" linguistic meaning is understood thespecific kind of content engendered by the process of reverberation in hu-man consciousness of objective, extralinguistic world, the latter forming theinner structure of linguistic units and with respect to which their expression,the sounds in which they are materialised, as it were, is the outer, phoneticstructure. As has been already mentioned there are two types of linguisticmeaning: the concrete lexical one and the abstract, grammatical.

The category of linguistic meaning cannot be understood or ex-pressed unless an insight is gained into the nature of linguistic expres-sion, the form, the vocal activities. Phonetics-phonology realised verysoon that in terms of meaning the sounds (and even phonemes) do notpossess or express meanings themselves, but they only help to differenti-ate meanings. By establishing the category of semiological relevance thephonetics-phonology has made the great stride that rapidly raised thisbranch of linguistics to the status of a theoretical discipline.

Accordingly investigations on the semantic level are to show howphonemes and morphemes are used to express meaning.

100

"The relations between language, thought and reality can well beillustrated by quoting from David Abercrombie: "The naive or commonsense view is that language reflects the world and our thinking about it,that to the categories of language correspond the categories of the realworld. Modern linguistics, however, inclines to the view that language isnot a passive reflection of, but rather an active practical approach to, theworld – a sorting out of it for the purpose of acting on it. Experience isdissected, split up, along lines laid down by nature. (The social bases oflanguage, 1965). Compare this judgement to what Benjamin Whorf wrotein his "Language, thought and reality": "Formulation of ideas is not anindependent process, strictly rational in the old sense, but is part of apar-ticular grammar and differs from slightly to greatly, between grammars.We dissect nature along lines laid down by our native languages....Theworld is presented in a kaleidoscopic flux of impressions which has to beorganised by our minds and this means largely by linguistic systems inour minds. We cut nature up, organise it into concepts, and ascribe sig-nificances as we do, largely because we are parties to an agreement toorganise it in this way – an agreement that holds throughout our speechcommunity and is codified in the patterns of our language." (Science andLinguistics, 1956, p. 460 – 468.) [1, p. 144].

There is very much along these lines which as yet remains unex-plained and has to be taken for granted. Exact historical information abouthow this or that unit or peculiarity of form and meaning had come intobeing, while others had been discarded, is in many cases not available.

But this does not mean at all that this is, in principle, inexplicable.On the opposite, language itself is cognizable, open to our cognition aseverything else around us. The facts of language are explained by theinterplay of a great number of factors which came in action jointly orseparately at different times and under different circumstances.

Thus, colour-terms are known in all human languages including red,the colour seen at least refracted end of spectrum, it is also said of shadesvarying from crimson to bright brown and orange, especially those seen inblood, sunset clouds, rubies, glowing coals, human lips and fox's hair; redis of bright warm colour, and as a general term is applied to many differentshades and hues, as crimson, scarlet, vermillion. At the same time peoplebelonging to different cultures may differently appreciate the colour refer-ences even, generally speaking, "sharing the same colour-system". Thus,for a Russian speaker of English red when said about somebody's hair willbe associated with reddish-yellow, while for an English speaker it willimply reddish-brown (or auburn). Interestingly, auburn came to Englishfrom Latin albus – white, light, this original meaning has changed over thecenturies to golden-brown or reddish-brown colour.

101

Another attempt in categorising linguistic units can be seen in wordswith -ster suffix, non-productive suffix in Modern English. There arewords youngster, gangster, spinster, monster that can be brought togeth-er as denoting a person belonging to a particular group (youngster – youth,gangster – a member of armed criminals, spinster – a woman who re-mains single after the convenient age of marrying), or being abnormalin size, shape, function (monster, gangster; perhaps spinster as violatinggenerally accepted conventions of social behaviour). In other words, theremay exist different dimensions of categorization and different groups,classes, systems established and singled out.

The scientific view that linguistic categories are not the same aslogical or psychological ones but highly specific categories to be ac-counted for by the cultural and historical background of the differentspeech communities was formed comparatively recently. The unpredict-ability and imaginativeness of grammatical categories with their specificbut always compulsory character are now generally acknowledged. Thenew outlook has also brought it home to the students of language that alllanguages are "classifying languages" with no exceptions made. The wordsof language are the ultimate components of sentences, which could notbe constructed if words did not fall into classes. These classes are oftendescribed as functional or "focus" ones. From a broader linguistic stand-point words of different classes are regarded as fulfilling particular syn-tactic functions because such are the properties of the class in question;the class being set up or constituted by a given set of grammatical mor-phological categories.

The words of a language are not categories, they have categories,and the similar categorial properties bring the words together into lexi-cal-grammatical classes of words called "Parts-of-Speech".

4.4. GRAMMATICAL CATEGORIES

A grammatical category is connected with a certain expression of acertain meaning which is systemically, recurrently rendered. To under-stand the process of establishing a grammatical category it is very impor-tant to introduce the concept of opposition. Grammatical categories arethe reverberations of the most general properties possessed by word-formsand lexical-grammatical classes of words as a whole.

In terms of categories a lexical-grammatical class of words may besaid to be well-determined, or even over-determined like verbs, or under-determined like nouns in Modern English.

102

A grammatical category is constituted by the opposition of no lessthan two mutually incompatible grammatical forms, like in: Let sleepingdogs lie, or It doesn 't stand even a dog's chance, where the noun inquestion is either used in the singular or the plural form, two making theCategory of Number; besides dogs and dog's appear as the realisationsof another category – of Case. This type of grammatical opposition iscalled a binary one.

There may exist in any class of words rudiments and relics of earlierstages of grammatical development. Thus, some nouns refuse to be used inthe plural (coffee, money), or in the singular form (news, means, trousers,scissors), they may take other grammatical inflexions for the plural: ox –oxen, goose – geese, child – children, analysis – analyses, etc.

A grammatical inflexional form carries at least one categorial formand thus of necessity belongs to some morphological grammatical cate-gory. In other words, in an inflexional form at least one grammaticalmorphological category is manifest. According to prof. A. A. Smirnitskythe minimal unit of grammatical meaning in its unity of minimal units ofcontent and expression is called a categorial form.

According to a definition of a grammatical category there must be noless than two categorial forms of the same category systemically and recur-rently rendered in a language, these categorial forms must be mutually in-compatible, that is mutually exclusive. They are dissimilar units of the mor-phological level that cannot appear simultaneously in an utterance. F. ex., in:We met to discuss our plans, – all units are syntagmatically dissimilar, butthey are not mutually opposed, we may take other class-members from nouns,verbs to make an utterance, like: Students met to discuss their plans, or evenStudents have already met to discuss their plans, etc. These changes are lessobligatorial, than the grammatical, paradigmatic ones. We cannot produceanything like Students has met, or We meets have discussed ... without vio-lating the fundamental rules of Grammatical Morphology and syntax. Anaction or an event either as singular or plural, past or present indefinite orcontinuous, real or problematic, etc.

The opposition of no less than two mutually incompatible categor-ical forms is the only possible realisation of a grammatical morphologi-cal category which exists only in the categorial forms, by means of them,through them. It would be wrong to assume that the term category can beused to refer to a categorial form, i. e.to denote a common case form, thesingular number form a category means to reduce the concept of catego-ry to that of a "generalised meaning".

The grammatical morphological category is of a passing nature; thenumber of the categories of a particular word-class, the number of cate-gorial forms can change in the course of time. The number of Case forms

103

in the English nouns is less today than it used to be in Old English, thenumber of oppositions in the English Verb system has drastically increasedsince that time.

Suppose, the independent use of 's to express possessivity and someother qualifications would become a recurrent growing and then domi-neering tendency, if so there would be no grammatical morphologicalopposition and the category of Case would disappear altogether, thus,compare: Mr Morrison's translation; All Balzak's characters; many ofthose R.A.'s Academy painters admitted to (Royal); uncle Jack's brother;Mary and Joan and Jane's room; The man I saw yesterday's daughter,etc., where the relation of 's to the qualified subject is becoming looser inevery given instance.

All categorial properties of a word as a representative of a particularlexical-grammatical class of words of a given language are shown (either overtlyor coverly) in each of its grammatical forms. But if a certain meaning can beimmediately observed in all forms of the same word, then this meaning is to beassigned to a certain lexical or lexical-grammatical category. The meaning ofthe repeated action as rendered by prefix re- in re-write, re-arrange is compat-ible with the meanings of all other grammatical forms of these words andcannot be included into the system of grammatical morphological oppositions.(Cf. in repeat the meaning of something said or done many times, or regularlybecomes fully lexicalised; Latin re – again and peto – to seek, also seen in:petition).

4.5. THE NOTION OF GRAMMATICAL CATEGORY

The notion of grammatical category (GC) has been widely discussedin a series of articles and manuals written by Russian linguists and hasalso been touched upon by representatives of various trends in linguis-tics outside Russia. Linguists, however, are often concerned with thedefinition of universal categories which, practically, reflect the unifor-mity of human thinking. However,no grammatical structure is universal,it differs from language to language.

No uniform opinion exists as to what units of language can be said torepresent exponents of grammatical categories. Can we apply this notionto, say, members of a sentence (subject, predicate, etc.) or to word-classes(noun, verb...)? What side of a linguistic sign does the notion cover, theplane of expression (form) or the plane of content (meaning), or both?

It should be kept in mind that in cognate languages there exist somestandard types of units, and interrelations between them, both on theparadigmatic and syntagmatic levels, that are common to all of them.

104

These can be studied on the basis of universal semantic and grammaticalcategories.

But while undertaking the exploration of the grammatical structureof a certain language, the definitions applied to universal categories haveto be adapted to the facts of the language under study, taking into ac-count the set of forms existing in it and the character of their interdepen-dence within the inventory of morphological exponents.

It is self-evident that such definitions as: "mood expresses the atti-tude of the speaker towards the action expressed by the verb"; "tenseindicates the time-point of the action"; "case indicates the relation ofnouns (pronouns...) to other words in a sentence", etc. are purely univer-sal and, therefore, inadequate to qualify the GCs and forms existing, forinstance, in English or any other language.

The functional analysis of linguistic units gives a clue to the solu-tion of some points connected with GCs because a functional approachimplies the registration of both paradigmatic and syntagmatic relationsof a linguistic unit or structure and takes into consideration form andmeaning in their interrelation. Such an approach signifies that beforedefining any grammatical unit or category in a given language one shouldbe aware of the specific arrangement of units in that very language.

A grammatical category of a language is a relational unit whichspecifies the meaningful relations between grammatical forms constitut-ing the peculiar, purely native structure of a language.

Relations between grammatical units are manifold because the unitsthemselves vary, therefore it would be only natural to expect that not allof them exhibit interdependence covered by the notion of grammaticalcategory. For example, the linear connections between the subject andpredicate of a sentence, the attribute and its antecedent, or those withinany word collocation, do not constitute the basis of grammatical catego-ry, such relations are based on a syntagmatic, linear concatenation andnot on a paradigmatic level.

The terms of a syntactic structure, such as e.g. an attributive group,have no identical unifying grammatical meaning common to both mem-bers of the collocation. For that reason syntactic relations and their unitsshould be excluded from exponents of a GC because it necessarily exhibitsuniform relations between grammatical forms where there is always somemоге or less general ground for comparison which serves to unite the termsof opposition or paradigmatic set on a semantic basis, the difference be-tween them (the terms) lying in the manner of marking certain distinctivefeatures (DF) – components of the plane of content of a form.

This argument seems to prove that a GC is a morphological notion.Within the sphere of morphology there exist parts of speech or word-

105

classes; established in accordance with the identity of semantic, formaldistributional characteristics. Speech parts are opposed to each other ofunits possessing contrasting grammatical features (formal and distribu-tional) serving as reason for their classification into different for thatmatter verbs, adjectives, nouns as word-classes cannot be called GCs.

The notion of GCs is confined to associative relations existing unitsof paradigmatic sets, to those general relations which unite all constitu-ents of a morphological paradigm. If, for instance, case is common to allnouns in German or Russian and a set of paradigms (or one paradigm)constitutes this grammatical notion, it can be regarded as a GCs becauseits exponents, different cases, are united by one common grammaticalmeaning and function, that of designating meaningful relations betweenwords of certain classes in speech. Mood, for example, is a GCs becauseit covers a certain set of special forms and their meaning individual inevery language, and so with every category. However, separate membersof a set constituting a GC and serving as exponents of a specific privatecategorial meaning cannot be called 'grammatical category’ because theyform part of some general, abstract GC, and not the category itself. Thus,the dative case (where it exists) is not a grammatical category, but can besaid to represent grammatical meaning, one of those which serve as com-ponents of a more general GC of case.

It follows from the criteria applied to the distinction of a GC that itcan be established exclusively on the basis of some feature in commonserving as a unifying factor, which is called 'ground for comparison', andat the same time each term of an opposition is characterized by the pres-ence or absence of a private grammatical meaning entering the semanticsphere of a broader GC.

It remains to be specified what side of a grammatical unit, memberof a paradigmatic set (which is necessarily bilateral, possessing both formand meaning), is covered by the notion of grammatical category, its ex-pression (form) or its content, or both.

The very fact that there can be established such notions as universalGCs signifies that a GC covers the plane of content of a morphologicalunit, which is the reason why it embraces only paradigmatic, associativerelations and not linear, syntagmatic. A grammatical category is definableonly on the basis of oppositions. The oppositional method presupposes theestablishment of the abstract categorial meaning unifying all the membersof a paradigmatic set, and separate semantic features, correlated with eachother, arising on the basis of a common general grammatical meaning.

This can be illustrated by the following example: let us state thatthe passive voice is a GC in English and that it finds expression in acertain set of forms such as is being done, was being done, etc. By means

106

of analysing the semantic structure of the forms containing one exponentin common, i.e. the combination of be + -en, which renders the meaningof passivity, we shall find that this meaning, as well as form, is commonto all the existing variations of this combination and thus serves as ageneral ground for comparison within the whole passive paradigmaticset. Oppositions in it will signal correlations of additional distinctivefeatures acting on the basis of a common passive meaning, as, for in-stance, in the correlation of the forms given above where the oppositionof present-past is evident both in meaning and form. Another feature incommon for the forms in the example above is duration.

Every grammatical meaning has its formal exponents adapted to thekind of semantic relations existing between grammatical units. Thereforethe grammatical DFs constituting a GC are reflected in a certain regular setof formatives which serves the purpose of expressing the structural organi-zation of morphological categories. These formatives are always regularand recurrent in parallel sets and forms. If we take, for example, the cate-gory of perfectness in English, covering the opposition between non-per-fect – perfect series of verbal forms, we can see that its exponents willhave the same formatives as well as an identical meaning in all sets of theverb: has done, had done, had been doing, will have been done, havingdone, to have done, etc. This fact confirms the regular form – meaninginterrelation, obligatory for all languages, which illustrates the regularityof the structural organization of any morphological system.

It has been maintained above that GCs of a language can be estab-lished after its morphological oppositions have been studied. For exam-ple, for case in English nouns the following definition can be offered:case as a grammatical category of the noun, expresses the relation ofpossession or belonging formally designated by the appended suffix -‘s(with its allomorphs) added to a certain class of nouns in the strong,marked member of the opposition and by its absence in the weak, un-marked member, devoid of any case-marker.

The English tense can be defined as a category expressing the timerelations of an action towards the moment of speaking and the moment ofutterance in the oppositions of past – present, present – future forms, andpresent – future, past – future-in-the-past. The same principle of defini-tions based on real, live oppositions is valid for any grammatical category.

In connection with the notion under discussion it is necessary todraw attention to the lack of uniformity among GCs. If a grammaticalcategory serves to denote the relations between two units arranged inlinear succession on the syntagmatic axis, such a category can be calleddependent. Dependent categories are: case (where it exists) and person-

107

number in the verb. The -s- ending in the third person singular, presenttense, is conditioned solely by the form of the subject of the action. Ex-ponents of such categories are easily reduced in the course of historicaldevelopment (as in English) because in linear juxtaposition similar rela-tions can be expressed in word order and functional elements, such asprepositions, pronouns. Such categories are syntagmatic by nature. B.Strang calls person and number "concord categories". Of quite a differ-ent nature is number in nouns. It is independent of its environment. Num-ber is of a concrete lexical character, it affects the referential meaning ofthe noun. Thus book and books denote different notions – one thing or amultitude of things. The most abstract, independent categories are tense,mood, as pect because they are inherent in the meaning of the verb as aclass of words denoting action, process: a process develops in time and itcan have some aspectual and modal quality. All these notions belong tothe process itself. The form of the subject or object does not affect theexpression of these verbal categories.

A semi-dependent category is voice. On the one hand, it denotes thesubject – object relations of a verb (its syntagmatic value), on the other,it preserves the abstract property of a verbal category which characteriz-es the action itself.

To generalize what has been said above concerning the GC it isnecessary to draw attention to the following:

• the notion of GC applies to the plane of content of morphologi-cal paradigmatic units;

• it refers to grammatical meaning as a general notion;• it does not nominate things but expresses relations, that is why it

has to be studied in terms of oppositions;• GCs of a language represent a realization of universal categories

produced by human thinking in a set of interrelated forms organized inoppositions;

• GCs are not uniform; they vary in accordance with the part ofspeech they belong to and the meaning they express;

• the expression of GCs in a language is based upon close interre-lation between their forms (exponents) and the meaning they convey.

The Grammatical Categories of the Verb (General Survey)After we have established what parts of speech in English possess

the ability to express GCs and what is a GC, it is logical to assume that inaccordance with the task set in this manual to concentrate our attentionon the verb morphology for the reason of its complexity and foremostimportance for the English morphology, we shall proceed with the de-scription of the meaningful categories expressed in the forms of the verb.

108

The following GCs find expression in the English verb forms:1) tense, denoting the reflection of objective divisions of time -

present, past, and future, plus, in English, time viewed from some pointin the past, so called future-in-the-past which we shall name 'future ІГ asopposed to 'future’ (simple future);

2) mood, expressing any supposition, non-fact (unreal mood(s) –subjunctive and conditional) as opposed to the expression of fact (indic-ative), or command (imperative);

3) voice, which denotes in the form of the verb that the subject ofthe action is acted upon (in the majority of cases), is not the agent of theaction in the passive voice, as opposed to the active voice;

4) aspect (duration), which marks the duration of the action in theform of the verb, as opposed to unmarked duration in the other memberof aspectual oppositions;

5) perfectness (temporal relativity), marking the anteriority of theaction to the temporal axis of orientation (present, past, etc.), as opposedto simultaneity of the action with the temporal axis;

6) person-number, which should be considered as a joint category,as it has one common exponent -s in the 3rd person singular, present, orfinds formal expression in the auxiliaries be and have; it stands outsidemeaningful oppositions in the verbal macrosystem.

The abstract GCs of tense (aspect), perfectness, and mood (active)constitute one upsplittable supercategory embracing the whole of the ver-bal vocabulary, its markers revolve in all its parts. Whereas the passivemicrosystem stands apart due to the limitation of the lexicon involved in it,and a marker (be + en) differing from those of the supercategory.

It would be natural to expect that the paradigmatic system of theverb adapted to the expression of GCs is structured according to somedefinite principles and presents a systemic organization of correlated el-ements, both in form, and in meaning, as GCs are always relational.

I.B.Khlebnicova. "Essencials of English Morphology" (p. 25 – 32)

109

SELECTED READER

H. SweetThe Practical Study of Languages. Grammar*

Grammar, like all the other divisions of the study of language, hasto deal with the antithesis between form and meaning.

The fact that in language there is generally a divergence betweenform and meaning – as when the idea of plurality is expressed by a vari-ety of forms, and sometimes by none at all (trees, men, sheep), or whenthe same form is used to express distinct grammatical functions (he seesthe trees) – makes it not only possible, but in many cases desirable, totreat grammatical form and grammatical meaning apart.

That part of grammar which concerns itself simply with forms, andignores the meanings of the grammatical forms as far as possible, is calledaccidence or "forms" (German formenlehre); that which concentrates itsattention on the meanings of grammatical forms is called syntax. Thusunder accidence an English grammar describes, among other details, thoseof the formation of the plural of nouns – how some add -s, some -es,while others mark the plural by vowel-change, and so on. In the syntax,on the other hand, the grammar ignores such formal distinctions as arenot accompanied by corresponding distinctions of meaning, or rather takesthem for granted, and considers only the different meanings and gram-matical functions of noun-plurals in general. The business of syntax is,therefore, to explain the meaning and function of grammatical forms,especially the various ways in which words are joined together to makesentences. As the form of a sentence depends partly on the order of itswords, word-order is an important part of syntax, especially when it servesto make such distinctions as in the English, The man saw the fox first,and The fox saw the man first. In fact, word-order is the most abstractpart of syntax, just as word-order is the most abstract grammatical form.

In accordance with its etymology, syntax is by some grammariansregarded entirely from this latter point of view, so that it is by them iden-tified with the analysis of sentences, the meaning of grammatical formsbeing included under accidence. Thus the peculiar meaning of the pluralinflexion in such words as sands, leads, waters of the Nile, would bysuch grammarians be discussed under accidence, on the ground that ac-

* for details see The List of Sources

110

cidence deals with isolated words, syntax only with combinations of wordsinto sentences. Although the application of grammatical terms cannot beallowed to depend on their etymology, yet, as we cannot avoid sayingsomething about the meaning of grammatical forms under accidence – ifonly to discriminate between such inflexions as trees, John's, comes – itis often convenient to clear off this part of the grammar under accidence,especially if the variations of meaning are only slight, or else so greatthat they cannot be brought under general rules.

The whole question is, after all, one of convenience. The separationof meaning from form is a pure matter of convenience, and is not foundedon any logical necessity, but only on a defect of language as it is, for in anideally perfect language form and meaning would be one – there would beno irregularities, no isolated phenomena, no dictionary, and what is nowdictionary and grammar would be all syntax. Even in languages as theyexist, form and meaning are inseparable, so that the separation of acci-dence and syntax must always be a more or less arbitrary one, which mayvary in different languages, quite apart from any questions of convenience.

We have seen that grammar deals with those phenomena of languagewhich can be brought under general rules, while the dictionary deals withisolated phenomena – especially with the meanings of separate words.

But not of all words. It is clear that while the meaning of such aword as man or house belongs to the dictionary, that of such a word as ofin the disobedience of man belongs to the grammar, for it has exactly thesame function as the -s of the genitive case: it cannot, indeed, be said tohave any meaning of its own at all.

From the point of view of the practical study of languages, such aquestion as whether or not the prepositions are to be treated of in the gram-mar as well as the dictionary, and the further question whether all of them,or only some of them, are to be included in the grammar, must be answeredby showing whether or not the acquisition of the language will be facilitat-ed thereby; and this will depend on the structure of each language.

We have seen that there is no real necessity for the separation of acci-dence and syntax. Although practical convenience often seems to call for aseparation, there may be circumstances under which it is desirable to treatforms and their grammatical functions and meanings together.

In this book I have also tried to do justice to another important prin-ciple of practical grammar, namely, that grammatical analysis has twostages, one of recognition or identification, and another of reproductionor construction. As I say in the preface, "The first requisite is to under-stand written texts, which involves only the power of recognizing gram-matical forms, not of constructing them, as in the further stage of writingor speaking the language."

111

I then go on to say, "All these principles are those which are carriedout – consciously or unconsciously – by most linguists. An experiencedlinguist in attacking a new language begins with the shortest grammar hecan find. He first takes a general bird's-eye view of the language, findsout what are its special difficulties, what has to be brought under generalrules, what to learn detail by detail, what to put off till a later stage. Therash beginner who starts with a big grammar forgets two-thirds of it soonafter he begins independent reading. Such a grammar as the one in thepresent work simply attempts to give him the really useful residue which,when once learnt, is not and cannot be forgotten."

The evils of the separation of syntax from accidence are well shownin the way in which the dead languages are taught in schools. Boys aremade to learn paradigms by heart, and are then set to read the classicalauthors with the help of a dictionary before they have acquired any realknowledge of the meanings of the inflexions they are expected to recog-nize in their texts – much as if they were taught the names of tools with-out being taught their uses.

It is now generally admitted that a grammatical rule without an ex-ample is of no practical use: it is an abstraction which is incapable ofentering into any direct associations with anything in the language itself.The example, on the one hand, is concrete: it can be imprinted firmly onthe memory by the mere force of the mechanical associations involved incarefully reading it and carefully pronouncing it aloud; while, on theother hand, it is logically associated with the rule, which it explains,illustrates, and justifies. The example serves also as a standard or patternby which the learner can recognize other examples of the rule as theyoccur in his reading. The example is thus a link between these otherexamples and the rule itself.

Many of the older grammarians, while expending much thought andcare on elaborating their statement of the rules, considered the choice ofexamples as of subordinate importance. They forgot that the first objectof grammatical study is not the acquisition of rules, but of a practicalcommand of the language itself; so that instead of the examples beingintended solely to illustrate the rules, the true relation is almost the re-verse: the rules are mere stepping-stones to the understanding of the ex-amples; so when the latter are once thoroughly understood, the rules be-come superfluous and may be forgotten.

(pp. 123 – 128)

112

J.R. FirthPapers in Linguistics 1934 – 1951

ContextualizationThe central concept of the whole of semantics considered in this

way is the context of situation. In that context are the human participantor participants, what they say, and what is going on. The phonetician canfind his phonetic context and the grammarian and the lexicographer theirs.And if you want to bring in general cultural background, you have thecontexts of experience of the participants. Every man carries his cultureand much of his social reality about with him wherever he goes. But evenwhen phonetician, grammarian, and lexicographer have finished, thereremains the bigger integration, making use of all their work, in semanticstudy. And it is for this situational and experiential study that I wouldreserve the term "semantics".

For the adequate description and classification of contexts of situa-tion we need to widen our linguistic outlook. Certain elementary catego-ries are obvious, such as speaking, hearing, writing, reading; familiar,colloquial, and more formal speech; the languages of the Schools, theLaw, the Church, and all the specialized forms of speech.

Then one might add such types of situation as those in which there isan "individual" or "monologue" use of language, and those in which thereis a sort of "choric" use, as when vocal interchange merely promotes ormaintains affective rapport. Malinowski has applied to this kind of linguis-tic behaviour the very happy phrase "phatic communion" - "a type of speechin which ties of union are created by a mere exchange of words".

Malinowski has also insisted on the specially interesting types ofsituation in which vocal interchange is just part of a job of work in hand,such as fishing, hunting, loading a truck, or the co-operative handling oftools and materials. He says the meaning of such words is "their prag-matic efficiency". Most of our contemporary "eye-language" in noticesand directions is of this kind.

A great deal of conversation or discussion may also be in prepara-tion for concerted or socially determined action. All the language of pub-lic administration and government may be said to be the language ofplanning and regulation, the language of public guidance. The subse-quent discussion of success or failure may be regarded both as "phaticcommunion" and as a situation in which something planned is either ac-complished or ends in failure.

In more detail we may notice such common situations as(a) Address: "Simpson!", "Look here, Jones", "My dear boy". "Now,

my man", "Excuse me, madam".

113

(b)Greetings, farewells, or mutual recognition of status and rela-tionship on contact, adjustment of relations after contact, breaking offrelations, renewal of relations, change of relations.

(c)Situations in which words, often conventionally fixed by law orcustom, serve to bind people to a line of action or to free them fromcertain customary duties in order to impose others. In Churches, LawCourts, Offices, such situations are commonplace. Your signature or yourword is a very important piece of linguistic behaviour. In passing, wemay notice that, when other things fail, judges often have recourse tovery rudimentary semantics in their interpretations. There is a great fieldfor practical semantics in the contextualization of crucial words injudi-cial remarks and judgements, particularly in the lower courts.

Such words are made binding by law, but many other words and phras-es are used with a similar binding effect in everyday life, because their usereleases overwhelming forces of public opinion, of social custom. "Be asport!", "I know you won't let us down". One of the magic words of the ageis plan. The mere use of this word and its derivatives releases certain forc-es of opinion and experience and gives the word weight. Its associationwith certain influential contexts gives it a power over us in this age ofuncertainty. Many more types of situation will occur to the interestedstudent,but there is an obvious need for a more accurate study of our speechsituations in order that categories may be found which will enable us toextend such social studies all over the world.

(pp. 27 – 31)

F.R. PalmerSemantics. A New Outline

Context of SituationThe term context of situation is associated with two scholars, first an

anthropologist who has already been mentioned, B. Malinowski, and latera linguist, J.R. Firth. Both were concerned with stating meaning in termsof the context in which language is used, but in rather different ways.

Malinowski's interest in language derived from his work in the Tro-briand Islands in the South Pacific. He was particularly concerned withhis failure to produce any satisfactory translations for the texts he hadrecorded. For instance, he recorded a boast by a canoeist which he trans-lated, "We-run front-wood ourselves ... we-turn we-see companion-ourshe-runs rear-wood." This, Malinowski argued, made sense only if theutterance was seen in the context in which it was used where it wouldbecome clear that, for instance, "wood" referred to the paddle of the ca-noe. Living languages must not be treated like dead ones, torn from their

114

context of situation, but seen as used by people for hunting, cultivating,looking for fish, etc. Language as used in books is not at all the norm; itrepresents a farfetched derivative function of language, for language wasnot originally a "mirror of reflected thought". Language is, he maintained,a "mode of action", not a "countersign of thought".

Malinowski's arguments were primarily based on his observation ofthe way in which the language of the people he was studying fitted intotheir everyday activities, and was thus an inseparable part of them. Buthe noted also that there is, even in our own more sophisticated society, aspecial significance of expressions such as How do you do? Ah, here youare, which are used to establish a common sentiment. [...] This aspect oflanguage he called "phatic communion", where the words do not conveymeaning but have a purely social function.

He noted, too that the child, right from the stage of babbling, useswords as "active forces" with which to manipulate the world around him.For the primitive man, similarly, words are "important utensils". Indeedfor him, Malinowski argued, there is much in common between wordsand magic, for both give him power.

Malinowski's remarks about language as a mode of action are usefulin reminding us that language is not simply a matter of stating information.But there are two reasons why we cannot wholly accept his arguments.First, he believed that the "mode of action" aspect of language was mostclearly seen in the "basic" needs of man as illustrated in the languages ofthe child or of primitive man. He assumed that the language he was consid-ering was more primitive than our own and thus more closely associatedwith the practical needs of the primitive society. To a very large degree,therefore, he assumed that the difficulties of translation were due to thedifferences in the nature of the languages and that the need to invoke con-text of situation was more important when dealing with primitive languag-es. But he was mistaken. For although there are the "primitive" people,who lack the knowledge and skill of civilised people, there is no sense inwhich a language can be regarded as primitive. Of course many languagesmay not have the vocabulary of modern industrial society, but this is areflection of the interests of the society, not of the primitive nature of thelanguage. In purely linguistic terms it appears to be a fact that no one lan-guage can be judged more primitive than another – though Malinowski isby no means the only scholar to make this false assumption.

The difficulties of translation that Malinowski noted result only fromthe differences between the languages, not the fact that one is more prim-itive. Secondly, Malinowski's views do not provide the basis of any work-able semantic theory. He does not even discuss the ways in which con-text can be handled in a systematic way to provide a statement of mean-

115

ing. Moreover, it is quite clear that even with his Trobriand Islandersmuch of their linguistic activity is not easily related to context. For in-stance, he discusses narrative, the telling of stories; but here, surely, thecontext is the same at all times – the story teller and his audience, what-ever the story. If context is to be taken as an indication of meaning, allstories will have the same meaning. Malinowski's solution was to invoke"secondary context". The context within the narrative; but that has noimmediately observable status and can no more be objectively definedthan concepts or thoughts that he was so eager to banish from discussion.

J.R. Firth, the first Professor of General Linguistics in Great Brit-ain, acknowledged his debt to Malinowski, but felt that Malinowski'scontext of situation was not satisfactory for the more accurate and pre-cise linguistic approach to the problem. For Malinowski's context of sit-uation was "a bit of the social process which can be considered apart" or"an ordered series of events in rebus" (i.e. an actual observable set ofevents). Firth preferred to see context of situation as part of the linguist'sapparatus in the same way as are the grammatical categories that he uses.It was best used as "a suitable schematic construct" to apply to languageevents and he, therefore, suggested the following categories:

A. The relevant features of the participants: persons, personalities(i) The verbal action of the participants.(ii) The non-verbal action of the participants.B. The relevant objects.C. The effects of the verbal action.In this way contexts of situation can be grouped and classified and

this is, of course, essential if it is to be part of the linguistic analysis of alanguage.

As an example of his use of context of situation Firth considered a"typical" Cockney event with the sentence:

"Ahng gunna gi' wun fer Bert""I'm going to get one for Bert.""What," he asks, "is the minimum number of participants? Three?

Four? Where might it happen? In a pub? Where is Bert? Outside? Orplaying darts? What are the relevant objects? What is the effect of thesentence?" 'Obvious!' you say."

It is important to stress that Firth saw context of situation as onepart of the linguist's apparatus or rather as one of the techniques of de-scription, grammar being another such technique on a different level, butof the same abstract nature. For linguistics was for him a sort of hierar-chy of such techniques all of which made statements of meaning. Herehe used the analogy of the spectrum in which light is dispersed into itsvarious wavelengths: linguistics similarly would "disperse" meaning in

116

a "spectrum of specialized statements. Thus, for Firth all kinds of lin-guistic description, the phonology, the grammar, etc., as well as the con-text of situation, were statements of meaning. Describing meaning interms of context of situation is, then, just one of the ways in which alinguist handles a language, and not in principle very different from theother ways in which he carries out his task.

Firth's views have often been criticised or even rejected outright,but the criticisms have usually failed to understand precisely what Firthwas trying to say. It will be worth while, therefore, to consider some ofthem since this may make Firth's standpoint clearer.

First, it has often been said that he was guilty of equivocation in hisuse of the word "meaning". For while context of situation may well dealwith meaning in the usual sense, i.e. the "semantic" sense, quite clearlythe other levels, grammar, etc. are not concerned with meaning in thesame sense. In claiming, therefore, that all the levels are statements ofmeaning and that context of situation was thus just one of a set of similarlevels, Firth was, consciously or unconsciously, using "meaning" in twodifferent senses, one legitimate, the other his own idiosyncratic usage.

This criticism is not entirely fair for three reasons. First, it is validonly if we accept that there is an area of linguistic investigation whichdeals with the relation of language and the world outside that is quitedistinct from the investigation of the internal characteristics of language.But, as we have already seen, many linguists have confined semantics tosense relations; for them at least, the study of meaning does not differgreatly in kind from grammar, since both would seem to be intralinguis-tic. I do not accept this point of view – I merely point out that Firth is byno means alone in seeing the study of meaning in the narrow semanticsense as not different in principle from the study of grammar. Secondly,we have already seen in the discussion of sense and reference, that it isalmost certainly impossible, in principle, to decide what is "in the world"and what is "in language". If this is so, Firth is surely to be praised ratherthan criticised for refusing to draw a clear distinction within his levels ofdescription between the one that deals with language and the world andthose that are wholly within language. Thirdly, Firth did not produce anytotal, "monolithic", linguistic model which could, in theory at least, to-tally describe a language. He did not, in fact, believe that such a modelwas possible even in principle (though nearly all linguists have assumedthat such a model is not merely possible, but essential). The linguist forFirth merely makes partial statements of meaning, saying what he canabout language where he can, cutting into it at different places like cut-ting a cake. There is no need on such a view to distinguish between state-ments that are about meaning and those that are not.

117

A second criticism of Firth's view is that it has very limited valuesince it will not get us very far. Context of situation may be all right forthe Cockney example or for the drill sergeant's Stand at ease, but not forthe vast majority of the sentences that we encounter. But this does notprove that Firth was wrong. If we cannot get very far with context ofsituation this is perhaps no more than a reflection of the difficulty ofsaying anything about semantics, and it is surely better to say a little thanto say nothing at all. It must be remembered too that Firth believed wecould never capture the whole of meaning. The proper conclusion, per-haps, should be that we need far more sophisticated techniques for con-text of situation than have yet been developed.

It is easy enough to be scornful, as some scholars have been, ofcontextual theories and to dismiss them as totally unworkable. But it isdifficult to see how we can dismiss them without denying the obviousfact that the meaning of words and sentences relate to the world of ourexperience. One virtue of Firth's approach was that he set out to makeonly partial statements of meaning. It may be that this is all we can everhope to achieve.

(pp. 46 – 54)

J. LyonsSemantics

What Is Communication?To say that language serves as an instrument of communication is

to utter a truism. Indeed, it is difficult to imagine any satisfactory defini-tion of the term "language" that did not incorporate some reference to thenotion of communication. Furthermore, it is obvious, or has appeared soto many semanticists, that there is an intrinsic connexion between mean-ing and communication, such that it is impossible to account for the formerexcept in terms of the latter. But what is communication? The words"communicate" and "communication" are used in a fairly wide range ofcontexts in their everyday, pre-theoretical sense. We talk as readily of thecommunication of feelings, moods and attitudes as we do of the commu-nication of factual information. There can be no doubt that these differ-ent senses of the word (if indeed they are truly distinct) are interconnect-ed; and various definitions have been proposed which have sought tobring them under some very general, but theoretical, concept defined interms of social interaction or the response of an organism to a stimulus.We will here take the alternative approach of giving to the term "commu-nication" and the cognate terms "communicate" and "communicative" asomewhat narrower interpretation than they may bear in everyday usage.

118

The narrowing consists in the restriction of the term to the intentionaltransmission of information by means of some established signalling-system; and, initially at least, we will restrict the term still further – tothe intentional transmission of factual, or propositional, information.

The principal signalling-systems employed by human beings for thetransmission of information, though not the only ones, are languages.[...] It will be assumed that the sense in which the terms "signal", "send-er", "receiver" and "transmission" are being employed in this section isclear enough from the context. [...]

A signal is communicative, we will say, if it is intended by the send-er to make the receiver aware of something of which he was not previ-ously aware. Whether a signal is communicative or not rests, then, uponthe possibility of choice, or selection, on the part of the sender. If thesender cannot but behave in a certain way (i.e. if he cannot choose be-tween alternative kinds of behaviour), then he obviously cannot commu-nicate anything by behaving in that way. This, we say, is obvious; andupon it depends one of the most fundamental principles of semantics -the principle that choice, or the possibility of selection between alterna-tives, is a necessary, though not a sufficient, condition of meaningful-ness. This principle is frequently expressed in terms of the slogan: mean-ing, or meaningfulness, implies choice.

"Communicative" means "meaningful for the sender". But there isanother sense of "meaningful"; and for this we will reserve the term "in-formative" and the cognate expressions "information" and "inform". Asignal is informative if (regardless of the intentions of the sender) it makesthe receiver aware of something of which he was not previously aware."Informative" therefore means "meaningful to the receiver". If the signaltells him something he knew already, it tells him nothing (to equivocatedeliberately with the verb "tell"): it is uninformative. The generally ac-cepted slogan, that meaningfulness implies choice, can thus be interpret-ed from either the sender's or the receiver's point of view. It is worthobserving, at this point, that sender's meaning involves the notion of in-tention and receiver's meaning – the notion of value, or significance. [...]

Under a fairly standard idealization of the process of communica-tion, what the sender communicates (the information put into the signal,as it were, by the sender's selection among possible alternatives) and theinformation derived from the signal by the receiver (which may be thoughtof as the receiver's selection from the same set of alternatives) are as-sumed to be identical. But there are, in practice, frequent instances ofmisunderstanding; and we must allow for this theoretically.

The communicative component in the use of language, importantthough it is, should not be overemphasized to the neglect of the non-

119

communicative, but nevertheless informative, component which is of suchimportance in social interaction. All utterances will contain a certainamount of information which, though put there by the speaker, has notbeen intentionally selected for transmission by him; and the listener willcommonly react, in one way or another, to information of this kind.

There are two further points having to do with the notion of com-munication which should be mentioned, though they will not be discussedin detail here. The first has to do with the distinction between the actualand the intended receiver of a signal. It is not uncommon for there to bemore than one receiver linked to the sender by a channel of communica-tion and for the sender to be communicating with only one (or somesubset) of these receivers. The sender may then include as part of thesignal some feature which identifies the intended receiver, or addressee,and invites him to pay attention to, or respond to, the signal. The mostobvious case of this in communication by means of language is when thesender uses a name or some other term of address in what we will laterrefer to as the vocative function. But the distinction between receiverand addressee is more widely relevant in communication, since, as weshall see later, the sender will often adjust what he has to say accordingto his conception of the intended receiver's state of knowledge, socialstatus, and so on.

The second point is of more general theoretical importance: that suc-cessful communication depends, not only upon the receiver's reception ofthe signal and his appreciation of the fact that it is intended for him ratherthan for another, but also upon his recognition of the sender's communica-tive intention and upon his making an appropriate behavioural or cognitiveresponse to it. This has long been a common place of non-philosophicaltreatments of meaning and communication (e.g., Gardiner 1932); and ithas been forcefully argued more recently, from a philosophical point ofview, by such writers as Grice (1957) and Strawson (1964).

As far as statements of fact (or what purport to be statements offact) are concerned, it is generally the case that the sender will intendthat the receiver should believe what he is told: that he should hold it tobe true and should store it in memory as a fact. Furthermore, the sender'sdesire to convince the receiver that such-and-such is true commonly de-rives from, or is associated with, some other purpose. For example, thereare all sorts of reasons why we might wish to draw someone's attentionto the fact that it is raining: we may think that he will be pleased to knowthat he need not water the garden; we may be concerned that he shouldnot forget to take his raincoat or umbrella; we may want him to close thewindow or bring in the washing. The particular purpose that we have intelling someone that it is raining will vary, but there will usually be some

120

purpose over and above our desire to inform him of a fact of which hewas previously ignorant. Indeed, it may be the case (and it commonly is)that what we actually say is of itself uninformative, in that the receiverknows (and we may know that he knows) whatever fact it is that we aredrawing to his attention. This does not invalidate in any way the notionsof communication and information with which we are operating here.

There is nothing paradoxical in the suggestion that a non-informa-tive utterance should be produced with the intention, that the receivershould infer from it (and from the fact that, despite its banality, it is ut-tered) something that is not said and in the context need not be said. Itmay be assumed, however, that the interpretation of non-informative ut-terances trades upon our ability to interpret the same utterances in con-texts in which they would be informative; so too does our ability to inferthe very specific and context-bound purposes that the sender might havehad for producing such-and-such an utterance-token on some particularoccasion. The sentence "It's raining" has a certain constancy of meaningwhich is independent of the specific purposes that someone might havein uttering it. The question is whether this constant meaning of "It's rain-ing" and of any arbitrary sentence that might be uttered in order to makea statement of fact can be said to depend intrinsically upon some moregeneral notion of communicative intention.

We will not go into this question here. Meanwhile, it may be point-ed out that, whereas it is clearly not essential to the notion of making acommunicative and informative statement that the person making thestatement should be speaking what he believes to be the truth or shouldintend the addressee to believe what he is being told, these are arguablythe conditions under which the communication of factual information isnormally assumed to operate.

(pp. 32 – 35)

R.H. RobinsJ General Linguistics. An Introductory Survey

Language and CommunicationMany definitions of the word "language" have been attempted and

they are to be found in dictionaries and in some textbooks. One definition,first set down in 1942, has enjoyed a wide currency: "A language is asystem of arbitrary vocal symbols by means of which a social group coop-erates." This definition covers much that is important, but in a sense alldefinitions are, by themselves, inadequate, hence, if they are to be morethan trivial and uninformative, they must presuppose, as does the one justquoted, some general theory of language and of linguistic analysis.

121

More useful at this point in an elementary book on linguistics will besome notice of certain facts that must be taken into account in any serious-ly intended theory of language. Language is, so far as we know now, spe-cies-specific to man. Every human being has acquired one language, hismother tongue, since childhood, the basic lexicon, grammar, and pronun-ciation within the first ten years of life, apparently without effort and with-out the inquirement of systematic instruction, in contrast to the actual teach-ing necessarily involved in the attainment of literacy and the mastery for-eign languages at school. Much that passes among conscientious parentsas "teaching a child to speak really amounts to the deliberate widening ofhis vocabulary along with his knowledge of the world.

The skills involved in speaking, being an acquisition taken for grant-ed and largely unnoticed in the process, excite no comment and evoke noadmiration; their absence in pathologically defective persons arousessympathy. We praise people for particular and relatively rare abilitiesthat depend on speech, for having a fine singing voice, for being a stir-ring preacher, an inspiring orator, or a good story-teller, and for beingable to recite with clarity a patter-song of the type written by W.S. Gu-bert, an unnatural exercise that taxes the powers of most otherwise fluentspeakers of a language. But all these accomplishments represent addi-tional abilities over and above the mastery of one's own first language.

Conversely, no other members of the animal kingdom have beenshown to possess anything like a human language. Of course animalscommunicate, and socially organized animals cooperate by means of vocaland other forms of communication. Much study has rightly been devotedto animal communication. Interestingly, the animal communication sys-tem in some respects nearest to human language (though a very long wayoff!) is the so-called language of bees, whereby bees that have been for-aging are able, by certain formalized movements often called "dancing",to indicate to other bees still in the hive the direction, distance, and rich-ness of a source of nectar, so that these others can make straight to it.This system shares with human language the ability to impart detailedinformation about matters not directly accessible to the senses of thosereceiving it; but we notice at once that the medium employed, the "sub-stance", as it is sometimes called, has nothing in common with the spo-ken medium in which all human language is primarily expressed.

Naturally studies in animal communication have centred on ournearest kin among the mammals, the primates, and specific investiga-tions have been made, for example, into the calls of gibbons in theirnatural habitat. But the area best known and most exciting to the generalpublic in this type of research has been the attempts to teach chimpan-zees to communicate with humans by human methods. Of these chim-

122

panzees, Washoe and Sarah, the subjects of prolonged training and studyin America, are the most famous. [...] Here it must suffice to point outthat attempts to teach chimpanzees actually to speak have largely failed;the signs used are in the main visual, involving gestures and facial move-ments. With this medium, intercourse involving information, questions,and requests, together with responses directly linked to them, and therudiments of syntactic structures, has made astonishing progress, far be-yond the scope of the language of bees, for example. But, and this is animportant reservation, bee language developed entirely within naturalcommunities of bees; chimpanzees have learned their language only af-ter prolonged association with human beings who have devoted them-selves to teaching them and studying them. Such studies tell us muchabout the inherent potentialities of chimpanzees, but they do not affectthe unique species-specificity of language in mankind. Human language,unlike every other communication system in the animal kingdom, is un-restricted in scope and infinite in extent. Against the severe restrictionsplaced on the topics about which bees and even trained chimpanzees cancommunicate, human beings will, in any language, talk about all the fur-niture of earth and heaven known to them and about all human experi-ence. Languages are adaptable and modifiable according to the changingneeds and conditions of speakers; this is immediately seen in the adapta-tion of the vocabulary of English and of other languages to the scientificand industrial developments, and the concomitant changes in people'slives, that took place in Europe and North America in the eighteenth,nineteenth, and twentieth centuries.

The immense power and range of language have been perceived inall societies, and the realization of them was, no doubt, partly responsiblefor the magical associations felt among some peoples to belong to certainwords relating to things and events vital to their lives or fearful in theireffects. Traces of such a magical outlook on language are to be seentoday in some familiar attitudes.

For all this flexibility and power, human languages have developedthrough the millennia in which mankind has existed on earth as Separatespecies through the medium of speech. The earliest known writing sys-tems do not date back more than about 4,000-5,000 years, a minute dis-tance in the time-scale of human existence. [...] all human language andeverything in human life that depends on language rests ultimately onthe distinguishable noises that humans are able to make out of the pas-sage of air through the throat, nose, and mouth.

Human infants inherit a biologically determined ability to acquireand use a language, and this inheritance may account for the universalfeatures found in all known languages and assumed in the rest; but we do

123

not inherit any particular language. A child learns the language of thosewith whom he is brought up in infancy and early childhood, whetherthey be, as is usually the case, his actual parents or others. There is nobiological preconditioning to acquire English rather than Malay or Ital-ian rather than Swahili.

Human progress is greatly hastened by the use of language in cul-tural transmission (one of its functions); the knowledge and experienceacquired by one person can be passed on to another in language, so thatin part he starts where the other leaves off. In this connection the impor-tance of the invention of printing can hardly be exaggerated. At the presenttime the achievements of anyone in any part of the world can be madeavailable (by translation if necessary) to anyone else able to read andcapable of understanding what is involved. From these uses of language,spoken and written, the most developed animal communication system,though given the courtesy title of language, is worlds away.

One topic connected with the study of language that has alwaysexercised a strong fascination over the general public is the question ofthe origin of language. There has been a good deal of speculation on this,usually taking the form of trying to infer out of what sort of communica-tive noise-making fully fledged languages in all their complexities grad-ually developed. Imitative exclamations in response to animal noises,onomatopoeia and more general sound mimicry of phenomena, excla-mations of strong emotion, and calls for help have all been adduced.Linguists, however, tend to leave this sort of theorizing alone, not be-cause of any lack of intrinsic interest, but because it lies far beyond thereaches of legitimate scientific inference, since we can have no directknowledge of any language before the invention of writing. In relation tothe origin of language, every known language is very recent.

Two frequently used analogies for attempted inference on the ori-gin of language are the acquisition of speech by children and the struc-tures and characteristics of so-called "primitive" languages. Both are in-valid for this purpose. Children acquire their native language in an envi-ronment in which language is already established and in constant andobvious use all around them for the satisfaction of needs, some manifest-ly shared by themselves. Their situation is entirely different from that ofmankind as a whole in the circumstances assumed to obtain while lan-guage itself was taking shape.

The second argument, based on the alleged nature of "primitive"languages, rests on a common, though deplorable, misconception of theselanguages. Linguistically, there are no primitive languages. There arelanguages of peoples whose cultures as described by anthropologists maybe called primitive, i.e. involving a low level of competence in the ex-

124

ploitation of natural resources and the like. Primitive, however, is not aproper qualification of language. Investigations of the languages of theworld do not bear out the assumption that structurally the languages ofpeople at different levels of cultural development are inherently differ-ent. Their vocabularies, of course, at any time reflect fairly closely thestate of the material and more abstract culture of the speakers; but lan-guages are capable of infinite adjustment to the circumstances of cultur-al development, and their phonetic and grammatical organization mayremain constant during such changes. It is a palpable fact of informedobservation in the linguistic study of the languages of culturally primi-tive peoples that phonetically and grammatically their languages are noless (and no more) systematic and orderly than the languages of WesternEurope and of the major world civilizations. Nor are the processes ofchange, that affect all parts of languages, any less active or any slower inoperation in these languages than in others; indeed, the converse may bethe case, as it has been held that the establishment of writing systems andstandards of correctness tend, if anything, to retard linguistic changes incertain situations. Every language has changes, irretrievably lost to knowl-edge, lying behind it. To argue from the language of primitive peoples tothe nature of a primitive stage in the evolution of language is valueless.

Languages fall into the class of symbol systems, symbols being aspecial class of signs. The science of sign and symbol systems, some-times called semiotics, lies outside the range of an outline introductionto general linguistics, but a brief clarification of the terms is desirable.Signs in general are events or things that in some way direct attention to,or are indicative of, other events or things. They may be related naturallyor causally, as when shivering is taken as a sign of fever, or as whenearthquakes are, or were, said to be signs of the subterranean writhing ofthe imprisoned god Loki; or they may be related conventionally and soused, and they are then called symbols, as, for example, the "convention-al signs" for churches, railways, etc. on maps, road signs, and the coloursof traffic lights.

Among symbol systems language occupies a special place, for atleast two reasons. Firstly, it is almost wholly based on pure or arbitraryconvention; whereas signs on maps and the like tend to represent in astylized way the things to which they refer, the words of a language re-late to items of experience or to bits of the world in this way only in theproportionately very small part of vocabulary called onomatopoeic. Theconnection between the sounds of words like cuckoo, hoopoe, and suchimitative words as dingdong, bowwow, rattattat, etc. and the creaturesmaking such noises or the noises themselves is obvious; and in a widerset of forms in languages a more general association of sound and type of

125

thing or event is discoverable, as in many English words ending in -ump,such as thump, clump, stump, dump, which tend to have associations ofheaviness, thickness, and dullness. It has been found experimentally thatmade-up words, like maluma and oomboolu, and takete and kikeriki, arealmost always treated alike by persons who hear them for the first timeand are asked to assign them to one or the other of a pair of diagrams, oneround in shape and the other spiky; the first pair are felt appropriate tothe former shape, and the second pair to the latter. The onomatopoeicand "sound-symbolic" part of language is of great significance, but itsextent in any vocabulary is quite small, and despite attempts by some tosee the origin of language in such imitative cries, it must be realized thatthe vastly greater part of the vocabulary of all languages is purely arbi-trary in its associations. Were this not so, vocabularies would be muchmore similar the world over than they are, just as the conventional pic-ture signs of several historically unrelated pictographic systems showobvious resemblances.

Secondly, what is conveyed by all other symbol systems can be ex-plained in language, and these other systems can be interpreted in lan-guage, but the reverse is not the case. The instructions given by road andrailway signals can be expressed in words, the propositions of logic canbe translated into ordinary language, though with loss of brevity andprecision, those of classical Aristotelian logic fairly directly, those ofmodern symbolic logic more indirectly. But in languages we deal withwhole areas of human life and engage in modes of communication withwhich logical systems as such have no concern.

(pp. 5 – 15)

W. HaasLinguistic Relevance

Linguistic AnalysisTo the linguist, fundamentally the most puzzling characteristic of

language is its enormous productivity - that obvious ability we have,every one of us, of always adding something new to an infinite variety ofutterances; of saying what has never been said before, and understandingwhat we have never heard before. To explain how this is possible, is theroot-problem of linguistic analysis. It is of course immediately clear thatthere are, and must be, certain partial similarities between the new utter-ances and the old – certain regularities in those operations of saying andunderstanding: what is novel must be assumed to arise from a new butregular combination of old parts. The first task, then, of linguistic de-scription – somewhat analogous о the chemist's attempt to "explain" the

126

puzzling variety of material things – may be said to be a reductive one:the task, namely, of reducing the practically infinite variety of utterancesto a relatively few current relations of a relatively few recurrent ele-ments. So much seems to be obvious and uncontroversial. But much thatis less obvious seems to follow from having located the problem anddetermined the task in this way.

Something about the objects to be described and analysed must betaken for granted; and the question is, how much. One may ask, for in-stance, whether it would be profitable to apply linguistic methods to ar-bitrary stretches of speech, i.e. to nothing more than certain events ofacoustic disturbance. We seem to require more, as a basis for linguisticoperations: nothing less, in fact, than that succession of different signif-icant utterances which is our problem. They are the objects of our anal-ysis. Precisely as the chemist is not concerned to verify the perceivedvariety of material things (including the instruments of his laboratory),so the linguist simply observes the various meaningful utterances andknows of their distinct existence, without feeling obliged to explain howhe knows it. Neither the chemist's nor the linguist's presupposition isbeyond question. But the question is asked by other disciplines – by psy-chologists, anthropologists, philosophers. The linguist asks about the in-ternal structure of his "facts", always taking for granted that he can ob-serve and roughly distinguish them.

J.R. Firth refers here to a "basic postulate", which he calls "the im-plication of utterance". "Language text," he says, "must be attributed toparticipants in some context of situation," before it can be analysed. Thevarious global forms and meanings of the pieces of text, which are sub-jected to linguistic analysis, are regarded as "given" by the linguist, thoughthey are capable of being "established" and "explained" by anthropolog-ical inquiries - that is, by a study of speakers and listeners in "contexts ofsituation". This tells us something about the relation between the linguis-tic and the social studies of human speech. "Context of situation," saysFirth, "makes sure of the sociological component" of linguistic descrip-tions. It is "a convenient abstraction at the social level of analysis, andforms the basis of the hierarchy of techniques", which are employed bylinguistic analysis proper.

L. Bloomfield was equally explicit on what he called "the funda-mental assumption of linguistics". "We must assume," he said, "that inevery speech-community some utterances are alike (or partly alike) inform and meaning." This is the basis of linguistic analysis. Bloomfield,however, made it unnecessarily difficult for himself, and his followers,

127

to feel at ease with this fundamental assumption. In Bloomfieldian lin-guistics, likeness of meaning becomes a source of trouble. It is formulat-ed in such a way as to allow mysterious "elements" of meaning ("se-memes") to attach themselves to the clear-cut formal elements of anygiven utterance. As a result, an unmanageable crowd of elusive extra-lingual entities intrudes, in haphazard fashion, into the very fabric oflinguistic analysis; whereas for Firth, situational context, lying itself safelyoutside the utterance, provides all we need to presuppose in order todescribe the significant functions and elements within.

It is on account of that "semantic" function of any described utter-ance is a function in "contexts of situation" – that Firth regarded of lin-guistic description, from syntax to phonetics, as a statement of various"modes of meaning". The unfortunate idiosyncrasy of this special termi-nology – which confined the term "semantic" to an anthropological study,while ascribing "meaning" to everything that is linguistically relevant,whether it be a sentence or a word or a mere sound - seems to have beenseriously misleading. It suggested a highly eccentric and esoteric theory,precisely at the point at which Firth in fact affirmed what is generallyacknowledged to be fundamental presupposition of linguistic studies.Indeed, he succeeded, where others had failed – namely, in admittingextralingual impositions, without allowing them to disrupt the properautonomy of linguistic studies. Extralingual "context of situation", whichall those sporadically intrusive "semantic entities", was simply thoughtof as infusing every linguistic element with relevance, or, as he put it,with some "mode of meaning".

Whatever the difference between Firth and Bloomfield, they agreedon one important point – namely, that the objects of linguistic analysisare not just physical objects, not arbitrarily selected stretches of speech".Such, they were sure, would not offer them any information for theirreductive task. "As long as we pay no attention to meanings," says Bloom-field, "we cannot decide whether guttered forms are "the same" or "dif-ferent"." To be able to declare such as this, we need to assume that someof those physical "stretches" make sense, and that they make differentkinds of sensefull information as to what specific sense any utterancemakes will not need to be presupposed as given). Meaningful utterances,then, many and various, are the objects of linguistic inquiry. The lin-guist's concern is with their internal constitution, the question being howto reduce them to recurrent elements in recurrent relations.

(pp. 116 – 119)

128

U. ECOThe Role of the Reader (Explorations in the Semiotics of Texts)

7.1. The Analysis of Meaning7.1.1. An intensional semantics is concerned with the analysis of

the content of a given expression. This kind of study has assumed in thelast two decades two forms, complementary and/or alternative to eachother: the interpretative analysis with the format of a compositional spec-trum of markers and the generative analysis in form of predicates andarguments. While the former approach seems to be exclusively concernedwith the meaning of elementary lexical entries, the latter seems to fit theneeds of a textual analysis which considers both the semantic and thepragmatic aspect of discourses.

I think, however, that such a clear-cut opposition should not be es-tablished. As it is proposed in Chapter 8 of this book, a sememe is initself an inchoative text, whereas a text is an expanded sememe. Theauthor who has more clearly advocated such an assumption (implicitlyas well as explicitly) is Charles Sanders Peirce. Some elements of Peirce'sthought can be reexamined in the light of such theoretical perspectives:Peirce's theory of interpretant cannot but lead to a form of analysis whichfits both the requirements of an interpretative generative semantics andonly from Peirce's point of view can the problems of contemporary texttheories be satisfactorily solved.

According to the principles of compositional analysis, a semioticexpression (be it a verbal item or any type of physical utterance) conveysaccording to linguistic conventions, an organized and analyzable it, formedby the aggregation (or hierarchy) of semantic features. Features consti-tute a system, either closed or open, and belong different contents ofdifferent expressions in different arrangements. Compositional analysisshould describe and define a virtually infinite of contents by means of apossibly finite ensemble of features, exigency of economy gives rise tomany aporias. It the features constitute a finite set of metasemiotic con-strue then their mode of describing a virtually infinite amount of its soundsrather disappointing. By such features as "human", late", "masculine", or"adult" (see Chomsky), one can distinct bishop from a hippopotamus,but not a hippopotamus from rhinoceros. If, on the contrary, one elabo-rates more analytic metasemiotic features such as "not-married" or "seal"(as it happens in interpretative perspective of Katz and Fodor), one isobliged to foresee an incredible number of other features such as "lion","bishop ог "with two eyes", therefore losing universality and running therisk that the set of metasemiotic features contains as many items as thelanguage to be analyzed.

129

Moreover, it is hard to establish which kind of hierarchy these fea-ture should be accorded to. A simple relation of embedding from genusto species can help only to a certain extent. It is, for example, obviouslyimportant to know that a schooner is a sailing ship, that a sailing ship isa vessel, a vessel – a boat, and a boat – a vehicle (marine), but this kind ofclassification does not distinguish a schooner from a brigantine, since itdisregards other features such as the form of the sails and the number ofthe masts. Provided this requirement is satisfied, it remains to be knownwhat purposes a brigantine or a schooner serves.

As a further criticism we can add that a compositional analysis interms of universal features does not say satisfactorily in which linguisticenvironments the item can be inserted without producing ambiguity. Thereare rules of subcategorization, establishing the immediate syntactic com-patibility of a given item, and there are selectional rules establishing someimmediate semantic compatibility, but these instructions do not go be-yond the normal format of a dictionary. Some scholars have proposed asemantic representation with the format of an encyclopedia, and this so-lution seems to be the only one capable of conveying the whole informa-tion entailed by a given term; but the encyclopedic representation ex-cludes the possibility of establishing a finite set of metasemiotic featuresand makes the analysis potentially infinite.

7.1.2. Other approaches have tried to overcome these difficulties byrepresenting the items of a lexicon as predicates with n arguments. Bier-wisch, for instance, represents father as "X parent of Y+ Male X 4- (An-imate Y + Adult X + Animate Y)" and kill as "X, cause (Xd change to(– Alive Xd – Animate Y)." This kind of representation not only takesinto account the immediate semantic markers (in form of a dictionary),but also characterizes the item through the relations it can have, withinthe framework of a proposition, with other items. In this perspective sin-gle semantic items are viewed as already inserted in a possible co-text.

Generative semantics has improved the use of predicate calculus, butshifting from the representation of single terms to the logical structure ofthe propositions (McCawley, Lakoff, and others). Only Fillmore has tried,with his case grammar, to unify both interpretive-compositional and gen-erative perspective. Fillmore remarks that the verbs ascend and lift areboth motion verbs and are both used to describe a motion upward, but liftrequires conceptually two objects (the one moving upward, the other caus-ing the motion), whereas ascendis a one-argument predicate. This remarkleads one to recognize that arguments, in natural languages, can be identi-fied with roles (similar to the actants in Greimas' structural semantics); forany predicate there is an Agent, a Counteragent, an Object, a Result, anInstrument, a Source, a Goal, an Experiencer, and so on. This kind of anal-

130

ysis solves very well the problem of the classification of features, follow-ing a sort of logic of action. Moreover, it satisfies the encyclopedic re-quirement and transforms a purely classificatory representation into anoperational schema: the composition of the meaning of a predicate tells ushow to act in order to give rise to the denoted action order to isolate itwithin a context. To walk, for instance, should mean that there is a humanagent, using ground as a counteragent, moving his body in order to dis-place it (as a result) from a spacial sourse to a spacial goal, by using legs asinstrument, and so on. However, some objections can be raised: (i) Where-as the roles can be recognized as a set of innate universals expressed by afixed inventory of linguistic expressions, the linguistic features which fillin these roles are again potentially infinite (how many kinds of instrumentcan be seen?). (ii) The proposal of such a "case grammar" seems to workapropos of predicates, but requires some additions as far as the representa-tion of arguments is concerned. Using a knife as instrument, I can killsomeone, but what about the semantic representation of knife? It seemsthat, more than a predicate argument structure, it could be useful in thiscase to employ such categories as who produces it, with material, accord-ing to what formal rule and for what purpose. This kind of representationrecalls the four Aristotelian causes (Efficient, Formal, Material, and Fi-nal); but the representation of an "object" could also be transformed intothe representation of the action required to produce this object (therefore:not knife but to make a knife), (iii) A complete semantic theory should alsotake into account syncategorematic terms such as preposition and adverb(for, to, below, while, and so on). According to the research of many schol-ars (Leech, Apresjan and others), it seems that this is possible, but we arefar from recognizing that those researches are to be considered both satis-factory and definitive. I think that an exploration into Peirce's theory ofinterpretant can strongly help to improve all these approaches.

7.1.3. There is, in any case, a sort of gap between contemporarycompositional analysis and Peirce's semiotic account of interpretants.Contemporary analyses are concerned mainly with a semantics of verballanguages, whereas Peirce was dealing with a general semiotics concern-ing all types of sign. I have elsewhere demonstrated that Peirce offers thetheoretical opportunity of extending the problem of compositional anal-ysis to every semiotic phenomenon, including images and gestures.

Nevertheless, in order to maintain a certain parallelism between thetwo poles of our inquiry, I shall limit the subject of section 7.2 to Peirceanproposals and examples concerning verbal language, even though thismethodological decision obliges me to underestimate the important rela-tionship between symbols, icons, and indices. Someone could object thatthis limitation is imposed by the very nature of my subject matter: Peirce

131

has said that only symbols (not icons and indices) are interpretable. "Prag-maticism fails to furnish any translation of meaning of a proper name orother designation of an individual object" (5.429); qualities have "noperfect identities, but only likenesses, or partial identities" (1.418). "Onlysymbols seem to be instances of genuine Thirdness (since they can beinterpreted), whereas icons are qualitatively degenerate and indices arereactionally degenerate, both depending on something else without anymediation (the icon from a quality, the index from an object)" (2.92 and5.73). Moreover, "it is not all signs that have logical interpretants, butonly intellectual concepts and the like" (5.482).

I think, however, that the context of Peirce's thought happily contra-dicts these statements. It is difficult to assume [...] that qualities are alwaysgeneral without asserting that they can and should be in some way definedand interpreted. And as far as icons are concerned, it should be remem-bered that the possibility of making deductions by observing those iconswhich are called diagrams depends on the fact that diagrams can be inter-preted and do arouse interpretants in the mind of their interpreters.

7.1.4. A sign-function correlates a given expression to a given con-tent. This content has been defined by a given culture irrespective of whethera given state of the world corresponds to it. "Unicorn" is a sign as well as is"dog". The act of mentioning, or of referring to, themes made possible bysome indexical devices, and "dog" can be referred to an individually exis-tent object, whereas "unicorn" cannot. The same happens with the imageof a dog and the image of a unicorn. Those which Peirce called iconic signsare also expressions related to a content; if they possess the properties of(or are similar to) something, this something is not the object or the state ofthe world that could be referred to, but rather a structured and analyticallyorganized content. The image of a unicorn is not similar to a "real" uni-corn; neither is recognized because of our experience of "real" unicorns,but has the same features displayed by the definition of a unicorn elaborat-ed by a given culture within a specific content system. The same can bedemonstrated apropos of indexical device.

The self-sufficiency of the universe of content, provided by a givenculture, explains why signs can be used in order to lie. We have a sign-function when something can be used in order to lie (and therefore to elab-orate ideologies, works of art, and so on). What Peirce calls signs (whichto somebody stand for something else in some respect or capacity) aresuch just because I can use a representamen in order to send back to afictitious state of the world. Even an index can be falsified in order tosignify an event which is not detectable and in fact, has never caused itssupposed representamen. Signs can be used in order to lie, for they sendback to objects or states of the world only vicariously. In fact, they send

132

immediately back to a certain content. I am thus asserting that the relation-ship between signifimtt and signifie (or between sign-vehicle and signifi-cation, or between sign and meaning) is autonomous in itself and does notrequire the presence of the referred object as an element of its definition.

Therefore it is possible to elaborate a theory of signification on thegrounds of a purely intensional semantics. I am not saying that an exten-sional semantics is devoid of any function; on the contrary, it controlsthe correspondence between a sign-function and a given state of the world,when signs are used in order to mention something. But I am stressingthe fact that an extensional semantics can be elaborated (and that pro-cesses of reference or mention can be established) only because an inten-sional semantics is possible as a self-sufficient cultural construct (that is,a code or a system of codes).

Can we say that the texts of Peirce entitle us to accept this perspec-tive? Obviously, in the Peircean framework, when signs are applied to con-crete experiences or haecceitates they are related to the indicated objects.But it is not by chance that Peirce established a difference between signand representamen; when he says that he uses the words "sign" and "repre-sentamen" differently, he means that the sign is the concrete, token ele-ment (the utterance) used in the concrete process of communication andreference, whereas the representamen is the type to which a coding con-vention assigns a certain content by means of certain interpretants.

"By sign I mean anything that conveys any definite notion of anobject in any way, as such conveyers of thought are familiarly known tous. Now I start with this familiar idea and make the best analysis I can ofwhat is essential to a sign, and I define a representamen as being whatev-er that analysis applies to. [...] In particular all signs convey notions tohuman minds; but I know no reason why every representamen should doso." I read this passage as the proposal of a difference between a theoryof signification and a theory of communication. Representamens are type-expressions conventionally correlated to a type-content by a given cul-ture, irrespective of the fact that they can be used in order to communi-cate effectively something to somebody.

Peirce continually oscillates between these points of view, but nev-er makes their difference explicit. Therefore when dealing with inter-pretants the object remains as an abstract hypothesis which gives a sortof pragmatic legitimacy to the fact that we are using signs; and when, onthe contrary, dealing with objects, the interpretant acts in the backgroundas an unnoticed but highly effective mediation which permits us to un-derstand signs and to apply them to such and such concrete experience.

(pp. 128 – 136)

133

W.N. FrancisThe Structure of American English

Building Blocks of Speech: Morphemics Morphs and AllomorphsThe linguist who has completed a phonemic analysis of a language

[,..] is in about the position a chemist would be in when he had succeededin isolating the elements. We have somewhat of an advantage over thechemist, for while he must keep a hundred and two elements, we haveonly 45 phonemes to worry about. But this doesn't help us a great deal.The number of possible combinations of our 45 phonemes is for all prac-tical purposes as great as the number of possible compounds of a hun-dred and two elements. There are so many, in fact, that only a smallpercentage of them are used in actual speech. Our next duty in studyingthe structure of English, therefore, is to see what combinations are used,and what they are like. The study of these matters is the province ofmorphemics. [...]

[...] we know that the phonemes by themselves have no meaning.Therefore, we conclude that the meaning must somehow be associatedwith the way the phonemes are combined. [...] Because these units haverecognizable shape, we call them "morphs", a name derived from theGreek word for "shape" or "form". A morph, then, is a combination ofphones that has a meaning. Note that each morph, like each phone, oreach person or each day, happens only once and then it is gone. Anothervery similar combination of very similar phones may come along rightafter it; if so, we will call this second combination another morph similarto the first one. If we are sure enough of the similarity, which must in-clude similarity of both the phones and the meaning, we can say that thetwo morphs belong to the same morph-type or allomorph. An allomorphcan thus be defined as a family of morphs which are alike in 2 ways: (1)in the allophones of which they are composed, and (2) in the meaningwhich they have. Or if we wish to be a bit more precise, we can define anallomorph as a class of phonemically and semantically identical morphs.[...] We may sum up the material of this section, then, as follows: A mor-ph is a meaningful group of phones which cannot be subdivided intosmaller meaningful units. An allomorph is a class of morphs which arephonemically and semantically identical; that is, they have the same pho-nemes in the same order and the same meaning.

MorphemesWith the recognition of the uniquely occurring morphs and their as-

sociation in sets of identical allomorphs, we have made a good start towardmoving up the ladder of linguistic structure to the next level. One thingseems certain even this early: we shall find a much greater number and

134

variety of units on this level than we did on the phonemic level. The num-ber of different combinations that can be made from 33 segmental pho-nemes is very large indeed. In fact, we can be sure that no matter howmany allomorphs we may discover, they will be only a small percentage ofthe total mathematical possibility. It is here, in fact, that the great diversityand adaptability of language begins to show itself. And it is here that wemust give up the hope of being as exhaustive in our treatment as we werein our discussions of phonetics and phonemics, that we cannot hope to listall the allo-morphs in English. Instead we can deal only with representa-tive types of illustrations of morphemic structure. [...]

As we have suggested in the title of this chapter, morphemes arebuilding blocks out of which the meaningful utterances of speech are puttogether. A morpheme is a group of allomorphs, each of which a combi-nation of phonemes: but, as we pointed out in the first chapter, in struc-ture of the kind the language shares with many other universal and man-made phenomena, the whole is more than the sum of all its parts. Whenphonemes are organized into an allomorph, meaning is added to make anew thing, just as when hydrogen and oxygen are organized into water, asubstance emerges that has new and different qualities which could nothave been guessed from a knowledge of the qualities of its components.From here on up the ladder of increasingly complex linguistic structure,we shall observe increasingly complex and precise indications of mean-ing, for after all it is to communicate meanings that language had beencreated, therefore, morphemes, the smallest structural units possessingmeaning, occupy a key position in linguistic structure. They are the fun-damental building blocks out of which everything we say is built.

Inflexion and DerivationSo far we have distinguished 2 principal types of morphemes: bases,

like [rat], and affixes, which are either prefixes, like [re-], or suffixes,like [-es]. Before we can proceed to the identification of words, which isthe ultimate goal of morphemics, we must look a bit more closely at thevarious types of affixes and the ways in which they occur. [...]

[We can differentiate between] two types of suffixes, a distinctionthat will be of considerable importance in our discussion of words, aswell as when we come to discuss grammar. [...] These suffixes whichmust always come at the end of the morpheme groups to which theybelong we will call inflexional suffixes. Those which may be followedby other suffixes we will call derivational suffixes. We can make a simi-lar distinction between the types of paradigms in which these suffixestake part. Thus a paradigm like [agree – agreed], the second form ofwhich consists of the stem plus the inflexional suffix [-ed], can be called

135

an inflexional paradigm, and the form [agreed ] can be called an inflect-ed form of agree. On the other hand, the pair[agree – agreement] illus-trates a derivational paradigm, and the form [agreement] is a derivativeform or simply a derivative of agree.

The suffixes of present-day English can thus be divided into 2 groups,inflexional and derivational suffixes. No such distinction exists in thecase of prefixes, however; they are all derivational. By means of prefixlike [dis-], for instance, a whole new set of derivatives of agree can bemade, corresponding to the derivatives already formed by adding suffix-es. In turn, these new derivatives may add inflexional suffixes so that wemay get such forms as ''disagreed", "disagreements", and "disagreeable-nesses". Since in adding suffixes all derivational ones must be added tothe base before the final inflexional one, we assume the same of prefixes.That is, inflexion takes place on a level of structure higher than that ofderivation. What this comes to is that, in terms of our examples, we treata form like ""disagreements" as consisting of [disagreement] + [-es], ratherthan [dis-] + [agreements]. Or, looking at it from the other direction, wemay say that in analyzing linguistic forms into their constituent mor-phemes, we separate inflexional suffixes first, before we separate deri-vational prefixes or suffixes.

Bound Bases. If we study such combinations as "conclude", "con-ceive", and "consist", we can observe that the stem of a derivative is notalways a free form; it may be bound. Thus, by comparing "conclude"with "occlude", "preclude", "include", and "exclude", we come to theconclusion that there is a morpheme [-klude], which serves as a stem forthese various derivational forms. Yet we never find it as a free form; thatis, we can find no environment into which [klude] fits in [...].

1. Bound morphemes are of 3 types: suffixes, prefixes, and boundbases.

2. Suffixes are either inflexional or derivational.a) Inflexional suffixes are always final in the morpheme groups to

which they belong. They are of wide occurrence, making large form-classes. Their distribution tends to be regular.

b) Derivational suffixes may be final in the morpheme groups towhich they belong, or they may be followed by other derivational suffix-es or by inflexional suffixes. They are of relatively limited occurrence,and their distribution tends to be arbitrary.

3. Prefixes are always derivational. Bound bases are morphemeswhich serve as stems for derivational forms but which never appear asfree forms.

(pp. 26 – 48)

136

Z. HarrisStructural Linguistics

MorphologyThe sequences (not necessarily contiguous) of phonemes or of com-

ponents which represent the flow of speech are now divided into newsegments each of which is uniquely identifiable in terms of phonemes(or components). This is done in such a way that each of these parts isindependent of the others in its occurrence over a stretch of any length(covering the whole utterance). The criteria for determining indepen-dence are selected in such a way as to yield a number of parts (morphe-mic segments or alternants), or rather the occurrences of such parts instated environments, are then grouped into classes (called morphemes)in such a way that all the members of a particular morpheme either sub-stitute freely for each other or are complements in corresponding sec-tions of the variant. Members of each morpheme can then constitute aclass called a morphophoneme.

We may therefore say that each morpheme is composed directly of asequence of morphophonemes, each of which in turn is a class consistingof one or more complementary phonemes or components. Each morphemehas only one morphophonemic constituency but the distinctions betweensounds are in general only in one-many correspondence with the distinc-tions between morphophonemes: two distinct morphophonemic sequenc-es may represent identical segment (or phoneme) sequences; such differ-ent morphophonemic sequences are phonemically equivalent.

It may be noted here that the morphemes are not distinguished di-rectly on the basis of their meaning or meaning differences, but by theresult of distributional operations upon the data of linguistics (this dataincluding the meaning – like distinctions between utterances which arenot repetitions of each other). In this sense, the morphemes may be re-garded either as expressions of the limitations of distribution of pho-neme, or (what ultimately amounts to the same thing) as elements select-ed in such a way that when utterances are described in terms of them,many utterances are seen to have similar structure.

The morphemes are grouped into morpheme classes, or classes ofmorphemes-in-environments, such that the distribution of one memberof a class is similar to the distribution of any other member of that class.

These morpheme classes, and any sequences of morpheme classeswhich are substitutable for them within the utterance, are now groupedinto larger classes (called position or resultant classes) in such a way thatall the morpheme sequences (including sequences of one morpheme) ina position class substitute freely for each other in those positions in theutterances within which that class occurs. All subsidiary restrictions upon

137

occurrence, by virtue of which particular member of one class or sub-class occurs only with particular members out of another, are stated in aseries of equations. The final resultant classes for the corpus, i.e. themost inclusive position classes, serve as the elements for a compact state-ment of the structure of utterances.

It is possible, however, to study other relations among the morphemeclasses than those of substitution within the utterance. The investigationof the relations between a class and sequences which contain it lead to ahierarchy of inclusion levels and to the analysis of immediate constitu-ents. The relations between one class and any other class which accom-panies it in an utterance may be expressed by long components of mor-phemes or of morpheme classes. And the investigation of substitutionwithin stretches shorter than a whole utterance leads to morphologicalconstructions and hierarchies of increasingly enclosing constructions.

The Criterion of Relevance: DistributionDescriptive linguistics, as the term has come to be used, is a partic-

ular field of inquiry which deals not with the whole of speech activities,but with the regularities in certain features of speech. These regularitiesare in the distributional relations among the features of speech in ques-tion, i.e. the occurrence of these features relatively to each other withinutterances. It is of course possible to study various relations among partsor features of speech, e.g. similarities (or other relations) in sound ormeaning, or genetic relations in the history of language. The main re-search of descriptive linguistics, and the only relation which will be ac-cepted as relevant in the present survey is the distribution or arrange-ment within the flow of speech of some features relatively to others.

The present survey is thus explicitly limited to questions of distri-bution, i.e. of the freedom of occurrence of portions of an utterance rela-tively to each other. All terms and statements will be relative to this cri-terion. For example, if the phonemic representation of speech is describedas being one-one, this does not mean that if a particular sound x is asso-ciated with a phoneme Y, then when we are given the phoneme Y weassociate with it the original particular sound X. The one-one correspon-dence means only that if a particular sound X in a given position is asso-ciated with a phoneme Y (or represented by the symbol Y), then when weare given the phoneme Y we will associate with it, in the stated position,some sound X, X, which is substitutable for the original X (i.e. has thesame distribution as X). In the stated position, the symbol Y is used forany sound which is substitutable for X, X’ etc.

In both the phonologic and the morphologic analyses the linguist firstfaces the problem of setting up relevant elements. To be relevant, these

138

elements must be set up on a distributional basis: X and Y are included inthe same element A if the distribution of x relative to the other element В,C, etc., is in some sense the same as the distribution of Y.

The distribution of an element is the total of all environments inwhich it occurs, i.e. the sum of all the (different) positions (or occurrenc-es) of an element relative to the occurrences of other elements.

Two utterances or features will be said to be linguistically, descrip-tively, or distributionally equivalent if they are identical as to their lin-guistic elements and the distributional relations among these elements.

J. BybeeAn Essay on Irrealis as a Grammatical Category

Grammatical Meaning in American StructuralismWhile it sometimes seems that linguistic theory and linguistic de-

scription can exist and develop independently of one another, every de-scription is based, at least implicitly, on a theory of language. This is justas true of descriptions of grammatical meaning as it is of descriptions ofgrammatical form. The influence of theory on the description of gram-matical meaning may not seem so very obvious, however, because ex-plicit attention to the nature and organization of grammatical leaning hashardly been the central focus of theoretical activity in this century. Al-though grammatical meaning has not attracted as much attention as syn-tactic theories, there have been major swings in the way it has been viewed,with consequent effects on the substance of grammatical description. Whatfollows is a brief summary of the major changes in such theories in Amer-ican linguistics.

The work of Boas, Sapir and Whorf on the grammatical systems ofNative American languages in 1920s and 1930's show grammatical mean-ing being treated with great sensitivity and respect. The exciting discoveryemerging from the newly-available data on a variety of native languages ofNorth America was that the concepts that require obligatory grammaticalexpression differ across languages, with some languages not using manyof the traditional categories of European languages, such as obligatorynumber or gender, and instead requiring clauses to carry information aboutevidentiality or temporal distinctions much more elaborated than those fa-miliar from European languages (Boas 1949: 206-207). A tension betweenthe emphasis on the similarities among languages and their differences isevident in the work of this period, with Whorf making his mark by arguingthat the Hopi tense/aspect/modality system is vastly different from any-thing imaginable in European languages (1938), while Sapir (1921) deftly

139

juggles differences and similarities among languages in working out a ty-pology of both morphological form and grammatical meaning. Still thedifferences among languages seemed profound, and it seemed natural thatsuch differences would reflect equally profound differences in the waysthat peoples of different cultures conceptualize reality.

This early start was not, however, followed by a great flowering ofinterest in grammatical meaning and its cultural and cognitive conse-quences. Instead American linguists turned away from the study of mean-ing to concentrate on the study of form, led by Leonard Bloomfield, whoalready in his 1933 book Language argued that as scientific investiga-tors, linguists have no direct access to meaning. The descriptive traditionthat followed Bloomfield attended very little to the meaning of grammat-ical morphemes in the languages being described. A greater emphasis onthe differences among languages arose. Beyond the categories of personand number, there seemed to be no hope that languages would carve upreality in similar ways, and a plethora of grammatical terminology arosefor categories of tense, aspect and mood, making such categories appeareven more different cross-linguistically. The autonomy of grammar frommeaning is asserted by Chomsky (1957), and finds expression in the de-scriptive tradition of tagmemics, in which morphemes are identified, notby meaning, but by its place in an elaborate numbering system whichindexes their distribution (an example is Turner [1958]).

The Influence of JakobsonWhile American Structuralism had taken a turn away from the con-

sideration of grammatical meaning and any possible form-meaning cor-relation, Roman Jakobson, in an effort that spanned five decades, articu-lated a rigorous theory of grammatical meaning based on structuralistprinciples. Although not many descriptions exist that adhere strictly toJakobson's theory, some of his principles have becamе so basic to lin-guistic thought that they have been assumed uncritically in descriptionsof both generative and more traditional learning. The most important ofthese principles is that of the semantic division, which gives rise to des-ignators of grammatical meaning, "past/nonpast", "future/nonfuture" and"realis/irrealis". The notion of opposition has several consequences forthe analyses of grammatical meaning. First, Jakobson proposed that allgrammatical oppositions were essentially binary and categories havingmore than two members could be analyzed with sets of binary features.This proposal was based on a firm belief that the binary oppoion repre-sents a logical operation very basic to human cognition and is further-more essential to language in that it simplifies multilateral oppositions(Jakobson 1972 [1990]).

140

Second, it follows from the notion of opposition that a grammaticalmorpheme (henceforth abbreviated to gram) takes its meaning the sys-tem of oppositions to which it belongs. Thus a present tense in a lan-guage that also has a past and a future will be different from a presenttense in a language that has a past but no future. In this view, grams donot have inherent meaning, but rather are defined by their relation toother members of the opposition.

Third, the categories of a Jakobsonian grammar are Aristotlean:boundaries between members of the category are discrete and the fea-tures defining the members are necessary and sufficient conditions. Thusthe semantic space covered by a gram is homogeneous – each occurrenceof the gram represents its features of meaning equally as well as anyother occurrence.

Fourth, each gram has one abstract, invariant meaning, a meaningthat is present in all contexts of use. Additional nuances or variations inmeaning are attributable to items in the context and are not part of themeaning that is derived from the sets of binary features defining the mean-ing of a gram.

These principles have found their way into the set of assumptionsthat linguists use when approaching the analysis of the grammatical sys-tem of a language, and they show up to varying degrees in descriptivework. For instance, it is common to see labels such as "past/non-past"and "future/non-future" used in grammatical descriptions, or even moreexplicitly binary features [+/- past], [+/-future], [+/-continuous], etc. (e.g.Li 1973) in place of fuller descriptions of the range of use of grams.

The assumption that a gram has one abstract meaning that is mani-fest in all its occurrences is, of course, an assumption that is essential tolinguistic analysis: one could not discern the meanings of morphemes,either lexical or grammatical, without assuming that they are constantacross conditions. Only in this way can one discover the cases in whichmeanings are not constant. It is the treatment of meanings that do differin context, e.g. the use of the English Past Tense in //-clauses yielding ahypothetical, but not past, sense, that is controversial. A Jakobsoniananalysis would insist that English Past Tense cannot mean "past" butrather must mean "remote from present reality" since it is used in situa-tions such as hypothetical ones, which are not past (Steele 1975, Lan-gacker 1978).

On the other hand, two other tenets of Jakobson's theory are notgenerally applied descriptively. First, Jakobson considered grammaticalmeaning to apply only to obligatory categories (Jakobson 1972 [1990])

141

thus excluding from his theory of grammar non-obligatory items such asauxiliary constructions, particles and derivational affixes. This exclu-sion leaves only a small core of grammatical categories to be analyzed,and indeed, in some languages, none at all. Most descriptions, to be com-plete, must also attend to the non-obligatory but still grammaticized itemsand constructions. In fact, many descriptions omit a discussion of oblig-atoriness altogether, perhaps because it is very difficult in many cases todecide whether or not a category is obligatory.

Another aspect of Jakobson's theory, which he regarded as of ut-most importance, but which has not been strictly adhered to in analyses,is the asymmetry between the members of a binary set. When two cate-gories are in opposition, one may signal the existence of a feature ofmeaning, but the other does not signal the absence of that feature; itsimply does not say whether the feature is present or not. Thus for Jakob-son, the negative value for a feature is always the unmarked value. Whilethe notion of markedness has pervaded all branches of linguistics, it hasalso generalized far beyond the strict definition assigned it by Jakobson.For most linguists today, the unmarked member of a category, or theunmarked construction type or interpretation is the one judged to be mostcommon and most usual, either in the language or cross-linguistically.

142

THE LIST OF SOURCES

1. Blokh, M.Y. A Course in Theoretical English Grammar /M.Y. Blokh. – Moscow: Vyshaya shkola, 1983. – 383 p.

2. Bloomfield, L. Language / L. Bloomfield. – London: Batsford,1955. – 419 p.

3. Chomsky, N. Essays on form and interpretation / N. Chomsky. –New York: North Holland, 1971. – 216 p.

4. Cruse, D.A. Some thoughts on agentivity / D.A. Cruse // J. ofLinguistics. – 1973. – № 1. – P. 11– 23.

5. DeLancey, S. Notes on agentivity and causation / S. DeLancey //Studies in Language. – 1984. – Vol. 8, № 1. – P. 181–213.

6. Dick, S.C. Functional grammar / S.C. Dick. – Amsterdam; NewYork: North Holland, 1979. – 230 p.

7. Fillmore, Ch.J. Regularity and idiomaticity in Grammatical Con-structions: the case of let alone / Ch.J. Fillmore, P. Kay, M. O’Connor //Language. – 1988. – Vol. 64, № 3. – P. 501–538.

8. Fries, Ch. The structure of English / Ch. Fries. – New York, 1957. – 200 p.9. Hjelmslev, L. Prolegomena to the theory of Language / L. Hjelmslev. –

New York: North Holland, 1953. – 216 p.10. Harris, Z. Method in Structural Linguistics / Z. Harris. – Chicago,

1960. – 134 p.11. Halliday, M.A.K. Halliday: system and function in language: select-

ed papers / M.A.K. Halliday. – London: Oxford Univ. Press, 1976. – 250 p.12. Ilyish, B.A. The structure of modern English / B.A. Ilyish. – Lenin-

grad: Prosveshchenie, Leningradskoe otd-nie, 1977. – 366 p.13. Jackendoff, R. Semantics and cognition / R. Jackendoff. – Cam-

bridge: MIT Press, 1983. – 283 p.14. Jespersen, O. The philosophy of grammar / O. Jespersen. – New

York, 1935. – 116 p.15. Katz, J.J. An integrated theory of linguistic description /

J.J. Katz, P.M. Postal. – Cambridge: MIT Press, 1964. – 178 p.16. Katz, J.J. Semantic theory / J.J. Katz. – New York: Harper &

Row, 1972. – 464 p.17. Langacker, R.W. Subjectification / R.W. Langacker // Cognitive

Linguistics. – 1990. – Vol. 1, № 1. – P. 5 – 38.18. Lyons, J. Introduction to theoretical linguistics / J. Lyons. –

Cambridge: Cambridge Univ. Press, 1968. – 519 p.19. Marchand, H. The categories and types of present-day English

word-formation: a synchronic-diachronic approach / Н. Marchand. –Wiesbaden: O. Harrassowitz, 1960. – 379 p.

143

20. McCarthy, M. English vocabulary in use / М. McCarthy,F. O’Dell. – Cambridge: Cambridge Univ. Press, 1994. – 296 p.

21. Muir, J. A modern approach to English grammar: an introduc-tion to systemic grammar / J. Muir. – London: Batsford, 1972. – 149 p.

22. Robins, R.H. General linguistics: an introductory survey /R.H. Robins. – 3rd ed. – London; New York: Longman, 1980. – 330 p.

23. Rosch, E. Human categorization / E. Rosch // Studies in cross-cultural psychology: in 2 vol. / ed. N. Warren. – London; New York,1977–1980. – Vol. 1. – 1977. – P. 1– 49.

24. Rosch, E.H. Principles of categorization / E.H. Rosch // Cogni-tion and categorization / ed.: E. Rosch, B.B. Lloyd. – Hillsdale; NewYork, 1978. – P. 27 – 48.

25. Rudzka-Ostyn, B. Semantic extensions into the domain of ver-bal communication / B. Rudzka-Ostyn // Topics in cognitive linguistics /ed. B. Rudzka-Ostyn. – Amsterdam; Philadelphia, 1988. – P. 507-553.

26. Schibsbye, K. A modern English grammar: with an appendix onsemantically related prepositions / K. Schibsbye. – 2nd ed. – London:Oxford Univ. Press, 1970. – 390 p.

27. Swan, M. Practical Еnglish usage / М. Swan. – 2nd ed. – Lon-don: Oxford Univ. press, 1995. – 658 p.

28. Sweet, H. The Practical Study of Languages / H. Sweet. – Ox-ford, 1964. – 174 p.

29. Sweetser, E. From etymology to pragmatics: metaphorical andcultural aspects of semantic structure / E. Sweetser. – Cambridge: Cam-bridge Univ. Press, 1990. – 174 p.

30. Talmy, L. How Language structures Space / L. Talmy // Spacialorientation: theory, research, and application: conf. on Spatial Orienta-tion and Perception, 1980 / Univ. of Minnesota; ed.: H.L. Pick, L.P. Acred-olo. – New York, 1983. – P. 225 – 282.

31. Talmy, L. The Relation of grammar to cognition / L. Talmy //Topics in cognitive linguistics / ed. B. Rudzka-Ostyn. – Amsterdam; Phil-adelphia, 1988. – P. 165 – 205.

32. Whorf, B.L. Language, thought, and reality: selected writings /B.L. Whorf. – Cambridge: Technology Press, 1956. – 278 p.

33. Wiezbicka, A. The semantics of grammar / A. Wierzbicka. –Philadelphia: J. Benjamins Publ., 1988. – 617 p.

34. Wood, F.T. The Groundwork of English Grammar / F.T. Wood. –London: Macmillan & Co, 1954. – 374 p.

144

RECOMMENDED LITERATURE

Ахманова, О.С. Фонология, морфонология, морфология /О.С. Ахманова. – М.: МГУ, 1966.

Бархударов, Л.С. Очерки по морфологии современного англий-ского языка / Л.С. Бархударов. – М., 1975.

Бархударов, Л.С. Структура простого предложения современ-ного английского языка , Л.С. Бархударов. – М., 1966.

Бархударов, Л.С. Грамматика английского языка / Л.С. Барху-даров, Д.А. Штелинг. – М., 1973.

Блох, М.Я. Вопросы изучения грамматического строя языка /М.Я. Блох. – М., 1976.

Блох, М.Я. Теоретические основы грамматики / М.Я Блох. – М., 2000.Бондарко, А.В. Грамматическая категория и контекст /

А.В. Бондарко. – Л., 1974.Воронцова, Г.Н. Очерки по грамматике английского языка /

Г.Н. Воронцова. – М., 1960.Гуревич, В.В. Теоретическая грамматика английского языка /

В.В. Гуревич. – М.: Флинта, 2003.Гуревич, В.В. Практическая грамматика английского языка /

В.В. Гуревич. – М.: Флинта, 2003.Долинина, И.Б. Синтаксически значимые категории английс-

кого глагола / И.Б. Долинина. – Л., 1989.Есперсен, О. Философия грамматики / О. Есперсен. – М., 1958.Иванова, И.П. Теоретическая грамматика современного англий-

ского языка / И.П. Иванова, В.В. Бурлакова, Г.Т. Почепцов. – М., 1981.Иванова, И.П. Вид и время в современном английском языке /

И.П. Иванова. – Ленинград: ЛГУ, 1961.Москальская, О.И. Грамматика текста / О.И. Москальская. –

М., 1981.Новое в зарубежной лингвистике. – М., 1981. – Т. X.Слюсарева, Н.А. Проблемы функциональной морфологии со-

временного английского языка / Н.А. Слюсарева. – М., 1986.Слюсарева, Н.А. Проблемы функционального синтаксиса со-

временного английского языка / Н.А. Слюсарева. – М., 1981.Смирницкий А.И. Морфология английского языка / А.И. Смир-

ницкий. – М., 1959.Солнцев, В М. Язык как системно-структурное образование /

В.М. Солнцев. – М., 1977.Сорокина, Т.С. Функциональная грамматика. Принципы линг-

вистического описания / Т.С. Сорокина – М., 1991.Трубецкой, Н.С. Основы фонологии / Н.С. Трубецкой. – М., 1960. A Communicative Grammar of English / G. Leech [et al.] – M., 1982.

145

A University Grammar of English / R. Quirk [et al.] – L., 1982.Baker, C.L. English Syntax / С.L. Baker. – MIT Press, Cambridge, 1995.Blokh, M.Y. A Course in Theoretical English Grammar /

M.Y. Blokh.– M., 2000.Brown, G. Discourse analysis / G.Brown, G. Yule. – CUP, 1996.Close, R.A. A Reference Grammar for Students of English /

R.A.Close. – Longman, 1990.Curme, G.O. The principles and practice of English grammar. Ap-

plied to present-day usage / G.O. Curme. – New York, 1957.Eco, U. The Role of the Reader (Explorations in the Semiotics of

Text) / U. Eco. – Bloomington: Indiana UP, 1984.Firth, J.R. Papers in Linguistics. 1934 – 1951 / J.R. Firth. – London, 1957.Francis, W.N. The Structure of American EnglishFrancis /

W.N. Francis. – N.Y., 1958.Fries, Ch. The structure of English / Ch. Fries. – New York, 1952.Greenbaum, S. An Introduction to English Grammar / S. Green-

baum. – Longman, 1991.Haas, W. Linguistic Relevance / W. Yaas // In Memory of J.R. Firth. –

London, 1966.Harris, Z. Structural Linguistics / Z. Harris. – Chicago, 1960.Hill, A. Introduction to linguistic structures / A. Hill. – New York, 1958.Hockett, Ch. A course in modern linguistics / Ch. Hockett. – New

York, 1958.Huddleston, R. Introduction to the Grammar of English / R. Hud-

dleston. – CUP, 1993.Ilyish, B. The Structure of Modern English / B. Ilyish. – L., 1971.Jacobs, R.A. English Syntax. A Grammar for English Language

Professionals / R.A. Jacobs. – OUP, 1995.Khaimovich, B.S. A Course in English Grammar / B.S. Khaimovich

, B.I. Rogovskaya . – M., 1967.Khlebnikova, L.B. Oppositions in morphology. As exemplified in

the English tense-system / L.B. Khlebnikova. – The Hague, 1973.Khlebnikova, LB. The Conjunctive Mood in English. As a problem

in general linguistics / L.B. Khlebnikova. – The Hague, 1976.Longman Grammar of Spoken and Written English / D. Biber [and

al.] – Longman, 1999.Lyons, J. Semantics / J. Lyons. – Cambridge, 1978. – V. 1.Nida, E. Morphology / E. Nida.– Ann Arbor, 1965.Palmer, F.R. Semantics. A New Outline / F.R. Palmer. – Cambridge:

CUP, 1977.Robins, R.H. General Linguistics. An Introductory Survey /

R.H. Robins. – London, 1964.Sweet, H. The Practical Study of Languages / H. Sweet. – Oxford, 1964.

Ответственный за выпуск: М.В. ВахмянинаКомпьютерная верстка: О.М. Нестерчук

Дизайн обложки: О.В. Канчуга

Подписано в печать 29.04.2009. Формат 60Ч84/16.Бумага офсетная. Ризография. Гарнитура Tаймс.

Усл. печ. л. 8,49. Уч.-изд. л. 9,60. Тираж 120 экз. Заказ .

Издатель и полиграфическое исполнение:Учреждение образования «Гродненский государственный

университет имени Янки Купалы».ЛИ № 02330/0133257 от 30.04.2004.ЛП №02330/0494172 от 03.04.2009.

Пер. Телеграфный, 15а, 230023, Гродно.

Учебное издание

МАЛЫШЕВА Ольга ЛеонидовнаМОЧАЛОВА Елена Олеговна

ТЕОРЕТИЧЕСКАЯ ГРАММАТИКААНГЛИЙСКОГО ЯЗЫКА

П о с о б и еВ 2 частяхЧасть 1

ISBN 978-985-515-152-5

9 7 8 9 8 5 5 1 5 1 5 2 5 >