This article was downloaded by: [Georgetown University]On: 26 September 2013, At: 12:52Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UK

Australasian Journal ofPhilosophyPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/rajp20

Proto-languageC.B. MartinPublished online: 02 Jun 2006.

To cite this article: C.B. Martin (1987) Proto-language, Australasian Journal ofPhilosophy, 65:3, 277-289, DOI: 10.1080/00048408712342941

To link to this article: http://dx.doi.org/10.1080/00048408712342941

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of allthe information (the “Content”) contained in the publications on ourplatform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views ofthe authors, and are not the views of or endorsed by Taylor & Francis.The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings,demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content.

This article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access

and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

A uslralasian Journal of Philosophy Vol. 65, No. 3, June 1987

P R O T O - L A N G U A G E 1

C. B. Martin

There was life before language in the history of the race and of the individual. The development o f the pre-verbal activity of the race and of the individual is by degrees a progression f rom the semantic ooze to sophisticated non- linguistic patterns o f inter-related behaviour, both shared and unshared.

The battle may well have been largely won 2 in showing that pre-linguistic and non-linguistic behaviour may be complex and intentional. I wish to take the battle still further against the persistent mystique o f language and argue that some non-linguistic activity at its more sophisticated and structured !e-'els has a remarkable pattern of parallels to that o f linguistic activity. It is a matter of degree, but when an agent shows enough of this pat tern o f parallels, this structured network of inter-related procedural activity can be called 'proto- language'. It should be remembered that the development o f linguistic behaviour is also a matter o f degree.

We can make a beginning toward seeing that non-verbal activity may, in itself, have semantic point by remembering that if we have not (by some supposition or assumpt ion or actual observation) settled on or hypothesised concerning an interpretation o f some other linguistic group's complex and inter-related non-verbal doings in and with their physical environment, there is no alternative non-magical route to the semantic point o f their sayings. One's knowledge of or supposition concerning another linguistic group's non- verbal doings is basic to and prior to and easier to ascertain than knowledge of their sayings. Watching a film in a language one does not understand should help to make that clear.

An example o f procedural activity described as proto-linguistic may seem over-interpreted only because the example is too undetailed. Any instance of procedural activity must be related to a vast complex of inter-related activity o f the agent for it to bear the suggested interpretation. Each case, then, would need at least the pages o f a novel or a full length film.

In the following account an intuitive range o f examples will be set out in enough detail so that they will be suggestive of the indefinite range o f inter- related behaviour required for the instances o f behaviour in the examples to have a degree o f determinateness o f semantic import .

An earlier version of this paper was read at the Center for the Study of Language and Information, Stanford University, September 20, 1985.1 wish to thank John Perry, Michael Bratman, John Etchemendy, Paul Snowdon, John Bigelow, John Heil, and Ali Kazmi for their helpful criticisms.

2 Cf. Robert Kirk, 'Rationality Without Language', Mind 76 (1967), Jonathan Bennett, Linguistic Behaviour, 1976, and C. B. Martin, 'The New Cartesianism', Pacific Philosophical Quarterly 4 (1984).

277

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

278 Proto-Language

The word 'language' is embedded in the word 'proto-language', but this is not meant to suggest that proto-language is a kind of language as sign language is a kind of language. It is meant to indicate that proto-language is a structured, rule-governed 3 network of semantic, procedural activity prior

to and basic to linguistic activity, having an almost totally unnoticed and surprising pattern of parallels to language itself.

This paper is a beginning toward detailing the structure and nature of proto- linguistic activity in its individual and social usage so as to provide a model for explaining a natural developmental flow (allowing also for and incorporating and making understandable the inventive 'leaps') in the semantic evolution of the race and the individual before language and into language. It is obvious that after (racially or individually) first acquiring a language, we do not and cannot abandon the use of the prOto-language. Their subtle inter-play and also their separate domains of semantic dominance and richness need investigation.

1. Word as t o k e n - w o r d as type. Sentence as token-sen tence as type.

2. The sentence, not the word, is the basic unit. R y l e - ' W e use words to m a k e sentences.'

la. Step as t o k e n - s t e p as type. Procedure as t oken -p rocedure as type.

2a. The procedure, not the step, is the basic unit. 'We use steps to make procedures. '

It is not easier to give clear and unproblematic non-intentional rules for the distinction between step and procedure than it is to give parallel unproblematic general non-semantic rules for the distinction between word and sentence.

There can be greater gaps between phonemes than between words in a sentence, so very mechanical general rules for word-detection are not possible. The semantic function or how the writing-the-token action or sounding-the- token action is used by the agent must be introduced and the rules for this are not clearly and unproblematically general. If semantics can be kept entirely out of syntax, no one has clearly shown how.

There are, of course, dictionaries that recognise and deal with words. There are also recognised steps for procedures. There can be instruction in non- verbal procedures showing discrete and sequential steps.

Before there were dictionaries there was instruction in the vocabulary. Before there were illustrated manuals there was instruction in the steps of various procedures- how to set the foot for a noiseless step in various hunting procedures. There can be pronunciation practice of a word and proto- pronunciation practice of a step to be used (practicing a grip on a golf club). The very same action used as a noiseless step in a stalking procedure could

3 The rules of proto-language, as also the rules of language, need not be explicitly known by a particular user or even by any user. Some of us are not sure that a clear description of what a rule of language is has been stated or that even one clear example of such a rule has been given.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 279

be given a different use as a non-slipping step in a walking across the ice procedure.

It may still be thought that there is a special difficulty in answering the question 'Why isn't any step in a procedure to be thought of as a procedure in itself?'

Almost any procedural steps could be used as a procedure just as almost any word could be used as a sentence. It depends, in both the linguistic case and the proto-linguistic case, upon how the action is used and seen by the agent. That is, whether the action is used and seen as a procedural unit for its own projected outcome or as only a part (step) in a total procedure taken as the unit, for a projected outcome. This is parallel to whether verbal action is u.sed and seen as a sentence unit for its own declarative use or as only a part (word) in a total sentence taken as the unit for its declarative use.

A particular action may have been used as a one word sentence or may have been used as a word in a sentence that got interrupted and so remains only an incomplete sentence.

Similarly, a particular action may have been used as a one step procedure or may have been used as a part of a procedure that got interrupted and so remains only an incomplete procedure.

3. Similar words in a sentence can have a different semantic use and different words in a sentence can have a similar semantic use.

3a. Similar steps in a procedure can have a different operational use and different steps in a procedure can have a similar operational use.

4. Similar sentences can have different semantic point (express different statements) in virtue of the similar words having a different semantic use and different sentences can have similar semantic point (make the same statement) in virtue of the different words having a similar semantic use.

4a. Similar procedures can have different operational point (perform different operations) in virtue of the similar steps having a different operational use and different procedures can have similar operational point (perform similar operations) in virtue of the different steps having a similar operational use.

5. One can ask 'Which sentence shall we use to express the statement?' 5a. One can ask 'Which procedure shall we use to perform the operation?' 6. The speaker has the capacity to make sentences expressing an indefinite

number of statements using a finite number of words. 6a. The agent has the capacity to make procedures performing an indefinite

number of operations using a finite number of steps. 7. Sa t i s fac t ion- the predicate of the sentence is satisfied by something. 7a. Sa t i s fac t ion- the d iscernings-about representational activity of the

observational procedure is satisfied by something. The non-linguistic procedural activity that I have called 'proto-language'

does not have anything like a full grammatical structure. It is not, after all, language. However, there is a rough parallel to subject-predicate that can yield a parallel to the satisfaction of a predicate.

A spatio-temporal region and something in it is represented by and fixed

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

280 Proto-Language

by, e.g., the direction in which one is looking or the position of one's hand in touching and direction of one's attention to what is within the region. This works roughly as the 'subject'.

By the selective, comparative, abstractive, contrastive, projective use of sensory input one performs procedurally discerning representational activity about what has been represented in the spatio-temporal region. This 'procedural-discernings-about representational activity 'works roughly as the 'predicate'.

This observational discernings-about-representational activity parallels (but, of course, is not the same thing as) the predicate of a sentence as the observationalfix of something parallels (but, of course, is not the same thing as) the subject of a sentence.

l have emphasised the non-behavioural activity, but what could be called 'experientially-loaded behaviour ' is commonly and importantly involved. One's own behaviour is experientially-loaded as one has visual, tactile, kinaesthetic sensory feed-back concerning that behaviour as one turns one's head for a better view, moves one's hand to get a better feel of the object, or walks closer to observe in more detail, etc.

8. A sentence is true iff the predicate is satisfied by something. 8a. An observational procedure is veridical iff the discernings-about

representational activity is satisfied by something. 9. What is or isn't (e.g., a rock falls down a cliff, does not hit a boulder)

that is represented in a sentence expressing a true statement, typically would still have been or not been though unrepresented.

9a. What is or isn't (e.g., a rock falls, etc.) that is represented in a procedure performing a veridical observational scanning operation, typically would still have been or not been though unrepresented.

10. What is required fo r there to be an unexpressed statement? There would be someone's real (though unexercised) linguistic capacity to

generate a sentence that would have or have had a particular semantic point (expressing the statement), and there is what is or isn't that would provide or would have provided satisfaction for the predicates if the individual had manifested its capacity to generate such a sentence.

10a. What is required fo r there to be an unperformed operation? There would be someone's real (though unexercised) procedural capacity

to generate a procedure that would have had a particular procedural point (performing an operation), and there is what is or isn't that would provide or would have provided satisfaction for procedural projection if the individual had manifested its capacity to generate such a procedure.

11. What is required fo r there to be an unexpressed statement without even the existence o f utterances or any individuals capable o f language?

It would be for there to exist a Way of using words simpliciter that could exist though nothing capable of using words existed.

11 a. What is required fo r there to be an unperformed operation without even the existence o f any individuals capable o f complex interrelated procedures?

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 281

It would be for there to exist a Way of using steps simpliciter, that could exist though nothing capable of using steps existed.

As there would be a Way of (non-existent) speakers using (non-existent) word-tokens, there would be a Way of (non-existent) agents using (non- existent) powdered unicorn horn.

Why not say instead that in such exigencies we are left just with what is or isn't, for that can be if realism is correct, without language or procedural action and even without beings capable of such action? Truth is a relation between such beings and what is or isn't. What is or isn't can be without truth, but truth cannot be without what is o r i sn ' t . 4

Such a stricture would keep faith with the truth that language consists in the capacities and actions of people.

However one goes in this contentious field, whether to luxuriate in universals or Ways or propositions or possible worlds, or to take the sterner stand that less is enough, the parallels between the linguistic and the non- linguistic continue. That is enough for the purposes of this paper.

12. The most basic kind of sentence is the declarative in which the sentence is a representation of something such that it is true or false, and the activity of representation does not serve primarily to affect what is represented.

12a. The most basic kind of procedure is observational in which the procedure is a representation of something such that it is veridical or non- veridical, and the activity of representation does not serve primarily to affect what is represented.

The representation of observing a rabbit running from a dog is achieved by the procedural use of sensory input. For example, it is by selectively attending to the movement of a rabbit, contrasting this against a stationary background of trees and mountains (and not attending to passing clouds) and comparative attention to the movement of the dog getting closer to the rabbit.

The status of such a simple case depends upon the degree to which it, as a pro cedure, fits for the agent into a complex interrelated network of such procedures. If the degree of fit is sufficiently sophisticated then even such a simple case could win the status of what I have called proto-language. All of this is required in making determinate the scope for veridicality and non- veridicality.

The agent in such observational cases need not be self-conscious about what it is using and how on the occasion of use any more than the normal speaker would be.

Different observational procedures can perform the same observational operation in virtue of different steps being given similar operational use.

For instance, selectively attending to the movement of the f ron t legs of the rabbit is a step in one's selective ordering and abstracting in the use of sensory input for the procedural representation of visually tracking a rabbit running f rom a dog. That step could have been replaced by the step of selectively attending to the movement of the back legs of the rabbit without altering the operational point of the representation.

4 cf. c. B. Martin, "Anti-Realism and the World's Undoing" Pacific Philosophical Quarterly, January 1984, for a defense of realism.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

282 Proto-Language

My account has not been couched in terms of belief states and desire states as intentional primitives. Let me explain why.

• . . universality belongs not to things themselves, which are all of them particular in their existence, even those words and ideas which are general • . . their general nature being nothing but the capacity they are put into by the understanding, of signifying or representing many particulars. (John Locke, An Essay Concerning Human Understanding, III-III-11.

We can paraphrase what Locke says by saying that nothing particular is intentional in itself but is made so only by its use. Beliefs and desires or ' information ' states are particular states and so in themselves are not intentional. Nor are they intentional because of their complex causal interrelationship with other states with perhaps bio-genetic explanations concerning their evolution through their complex environmental sensitivities. That account does well enough for such non-intentional states as states of the digestive system. 5

Following Locke, and employing his seminal notion of partial considering, intentionality comes only f rom the procedural uses by the agent of its states• Belief or desire states are not the kinds of things for such procedural use. They are instead capacity (and tendency) states for the use of what can be used. What can be used is our sensory input and imagery and our experientially loaded (through sensory feedback) behaviour.

It is more fruitful to consider the explication of belief and desire in terms of the doings and an interrelated complex of doing capacities (and tendencies) of which the experiential is an intrinsic and essential part or aspect of anything used in such a doing (and if the doing is a 'mental ' doing, it is the principal part), than it is to explicate such experientially loaded doings in terms of beliefs and desires•

This active, procedural model for perceptual discernings replaces the model of perception that either gives no role to sensation (replacing it with belief) or gives sensation only the dumb middle-man causal role of being an effect of physical stimuli and a cause of belief. In the active model I am suggesting it is the material for procedural use.

Paul Snowdon in correspondence suggested that only in the linguistic case one reason for going through the activity of saying that it is raining is that it is a way of making a declarative representation that it is raining. The characterisation is part of an explanation of the action.

In a parallel way, however, I can put myself in the observational position of having a good look and feel at what I already k n o w - t h e familiar contours, colours, texture, etc. of my own hand. Part of the reason for engaging in such representational activity is that it is a representation. I like the look and the feel which I know that I can alter (by half closing my eyes, etc.) without altering the hand. I am not doing this to get new information or even to re- affirm old knowledge. It may be as close as one can get to representation for representation's sake.

5 cf. c. B. Martin and Karl Pfeifer, "Intentionality and the Non-Psychological", Philosophy and Phenomenological Research 4 (1986).

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 283

I have tried to isolate activity pursued only for the semantic purpose of representing something to be the case. This purity is rare in both linguistic and non-linguistic activity, but I have shown it is possible to achieve as much in the non-linguistic as in the linguistic case.

On pages 148-49 of Jonathan Bennett's splendid book, Linguistic Behaviour, he states:

The human situation differs from the tribal one in another way. To explain it, I need to distinguish three ways in which U might get A to believe that P. (1) He might reveal to A some intention-free evidence for P, i.e. some item which is evidence for P independently of U's intentions in making it available to A. (2) He might employ the credence mechanism to get A to believe P, by some action containing what I call a 'natural pointer ' to P, that is, something which naturally tends to induce in the observer some thought closely connected with P; for instance, a pantomine of a snake if P is about a snake. (3) He might employ the credence mechanism by means of an action which did not naturally point to P, e.g. saying 'There is a snake near you' to warn A that there is a snake near him.

(2) and (3) could be described as involving, respectively, 'iconic' and 'non-iconic' vehicles of meaning. The 'natural pointer ' involved in (2) need not be 'iconic' in the strict sense of involving a resemblance between the action and something to do with P; for the pointer could be an actual object which is involved in P - f o r example, U might hold up a dish meaning that he wanted A to give it to him. Still, 'iconic' is a handy label, and I shall use it.

I do not think that Bennett's classifications capture the following kind of case.

Non-linguistic modes of mathematical thinking in topological and geometrical ways are achieved by projections f rom actual and imagined, performed and unperformed, manipulations and groupings of objects. It is procedural and operational thinking. This may be shared or not shared.

A carpenter may come to know non-linguistically the general characteristics of m a t e r i a l s - of different kinds of wood - what kind of wood is suitable for what kind of use, with constant projections (right or wrong) for new and untried kinds of use. It is procedural and operational thinking. This may be shared or not shared.

Someone who has mastered these ways of procedural projective thought may, on sighting a particular shape of a part of a distant mountain, use this in his or her various complex procedures of thought to project this shape as part of a plan for the possible use of a special wood to be used in a very special new fitting relationship in a novel shelter design.

Bennett has considered only the communicative behavioural forms of representation and this worked well for the purposes of his argument. I have been emphasising forms of representation that are not for, or not primarily for, communication. The shared and communicative uses of even language itself seem to me to have been over-stressed. At least eighty percent of our linguistic activity is neither shared nor communicated. It occurs in our silent

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

284 Proto-Language

verbalisations in the head and talking to ourselves, or notes written only for one's own eye. It is the most common mode of representation for our assessments, claims, wonderings, plans, projects, reflections and hypothesisings. Such activity deserves full status as linguistic utterance. It is not a kind of rehearsal for the 'real' t h i n g - i t is the real thing. It is not something 'behind' u t t e rance- i t is utterance.

The socialisation and sharedness of semantic activity both linguistic and nonqinguistic has been over emphasised and wrongly been made exclusive. (Indeed, the first linguistic use of an artificial sign could well have been the heaping of stones as a communicative act by someone to remind i tsel f at a later time of where it had buried something, or the breaking of branches to give i tsel f reminders of the way back to the point from which it started.) How far a child would go with its own unshared ideolect is a real question that has been botched by bad research.

There are things that an infant must begin to do and even continue in doing for itself.

An infant's sensory discriminations, comparings and contrastings, its reachings and graspings under sensory control are not all learned from others. Its own inventive way of crawling need not have been socially reinforced. Watch a child find its own way with things without sharing this with or getting reinforcement from anyone. Yet the child may retain these skilled procedures enjoying its success. No doubt the child as a social being would at least need what might be called 'hugs-in-general' that might reinforce all sorts of inventive activity to which it was not at all specially directed.

The 'born Crusoe' is quite another case and would have to be nurtured impersonally and most luckily by nature and it would probably have to be by nature itself (and most improbably so) the very limit case of the autistic. At least, it is no model for us, who need not only hugs-in-general, but a great many hugs-in-particular for our reinforcements.

We are, of course, social animals, but we are private animals too with private ways of learning and private ways of performing, some of them basic to the social ways.

Much of our procedural activity that is directed successfully or unsuccessfully to an outcome is not learned from or, especially in its many inventive forms, is not merely a socialised result of, interaction with others and it is not a shared activity let alone a conventionally shared activity and its purpose is not communication.

There are natural procedural ways of behaving directed to a particular outcome (that may or may not be successful) that need not be learned from others. These procedures are reinforced by success of outcome and not necessarily by reward from others. I believe that, to a substantial degree, we are innately built so that success is, to a degree, its own reward and we are, fortunately, innately built to bear up, not necessarily to be in all ways dependent on social encouragement and to repeatedly try against failure. The physical environment itself has a role to play in our learning by rewarding with success and punishing by failure our endeavours.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 285

I have meant to emphasise the important role of the individual's semantic enterprise in which it can carry on with what works, not needing others to 'catch on' and share before its action has semantic point. Surely, no one could think that immediate uptake was a requirement. It follows, then, that perfectly good semantic procedure can be used by an individual and lost for want of, or have a long wait for, effective sharing. Indeed, if semantic procedure needed an immediate capacity for uptake from others for sharing and communicative use, then the evolution of the proto-language and the verbal language would be made impossible if not incoherent.

This discussion of the private proto-language is not meant to touch that philosophical tar. baby, the private-language argument in any of its many forms, but only to dissolve what can't be gone around.

I wish to argue for a mixed economy. Let us allow a degree of private entrepreneural enterprise and a degree of socialisation in semantic endeavours.

13. The semantic point of a declarative utterance is not to cause a belief in someone else or to 'share' or to communicate, it is, instead, to go through an utterance-representation procedure that is true or f a l s e - t o represent successfully or unsuccessfully what is.

Frequently, one satisfies that semantic point in one's silent or solitary sayings or writings. This is not to deny that an utterance of a declarative sentence may not have many other functions as well as to perform its semantic point.

13a. The semantic" point of an observational representational procedure is not to do anything to show someone else or even to get anything done, but rather to represent successfully or unsuccessfully what is. This is not to deny that the performance of a representational procedure may not have many other functions as well as to perform its semantic point.

14. Communication by means of shared rule-governed verbal activity in which assent and dissent may figure.

14a. Communication by means of shared rule-governed procedural activity in which acceptance and rejection may figure.

People working together in the building of a complex structure may be doing more than birds building a nest. They may be cooperatively and inventively testing by various procedures the feasibility of other procedures, for the fit, the strength, the stability and the beauty of the various alternatives. And in this, one can correct or reject the procedural recipes of the other and substitute one's own. In that kind of complex and inter-related pattern of activity, one person rejecting a piece (perhaps as a sample of a kind of building material) that will not be suitable is its proto-parallel to dissenting to the offering-as-procedure. Its accepting a piece that is suitable is its proto-parallel to assenting to the offering-as-suitable procedure.

15. Sentences- recta-sentences. 15a. P rocedures - recta-procedures. Just as there can be sentences about sentences, so there can be procedures

about procedures. A testing procedure, perhaps with a model, for another procedure is a recta-procedure.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

286 Proto-Language

16. Sentence meaning--speaker meaning. To have the notion of the sentence m e a n i n g - speaker meaning distinction

requires that the individual should be capable of taking the sentence as an object of attention and assessment and not just what is represented by the sentence. It requires also that the individual should have some notion of an accepted or set way for the sentence to function (either in some shared, public use or even in the individual's own rule-governed ideolect) as well as of how the individual intended it to function which may or may not be at variance with the accepted or set way.

16a. Procedure m e a n i n g - agent meaning. To have the notion of the procedure m e a n i n g - agent meaning distinction

requires that the individual should be capable of taking the procedure as an object of attention and assessment and not just the outcome projected by the procedure. It requires also that the individual should have some notion of an accepted or set way for the procedure of function (either in some shared, public use or even in the individual's own rule-governed procedural network) as well as of how the individual intended it to function which may or may not be at variance with the accepted or set way.

An example of an accepted procedure would be that of a recipe (not in verbal form) developed and practiced by an individual for a particular dish.

One day, the individual follows that procedure but, by some mental lapse, takes it to be for a quite different dish.

All of the permutations possible in the linguistic case are paralleled in this procedural case.

17. One may say what one believes is not the case, usually in an attempt to deceive someone else concerning what is the case. In the course of making a declarative utterance concerning what one believes to be false may by mistake and against one's wishes say what is true. Or, having no beliefs one way or the other, one may utter a sentence constituting a guess that by luck is true or by luck is false. One can even by slip of the tongue say what is true whether one believes it or not. This is how one is led by the nose, as it were, by the language.

17a. One may perform a procedure projective of a particular outcome that one believes will not be the case. In the course of this, the outcome may be satisfied to one's surprise. O n e m a y believe that doing something- turning the key in the lock one final, exasperated time (not because one thinks it is worth trying, but out of anger) will not open the lock, and it does. The procedure has the right outcome against one's beliefs. Or, having no beliefs one way or the other, one comes to two paths and one guesses in the form of walking directedly down the path. This is a procedure projective of a particular outcome that only by luck is satisfied. One can even by a slip of the hand as part of one's procedure have an outcome satisfied whether one wants it or not.

It is interesting to see how natural it is for us to take credit without guilt for the undeserved truths of our linguistic declaratives or the undeserved satisfactions of outcomes of our procedures. Since we are charitable not only

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 287

to our own unintended linguistic and proto-linguistic successes but even to the unintended success of others, we are all in on the cheat of this particular inverted form of the principle of charity. This is understandable and even perhaps laudable given our general desire for the satisfaction of outcome of what we do and the truth of what we say, whatever form, intended or unintended, it may take.

18. In some cases of verbal declarative activity (such as saying 'That, over there, is a big one.') the perceptual perspective of the agent is used as a constituent of the declarative action. It is worse than misleading to call this the 'context' of the assertion. For the verbal declarative action to have the semantic content it has and to be representative of aparticular state of affairs, the perceptual perspective must be used by the agent as a constituent in the declarative action as much as the verbal sounds must be used as constituents.

18a. In some cases of procedural actions (such as getting water to be ready to douse a particular fire as opposed to having it ready for any fire that may occur) the perceptual perspective of the agent is used as a constituent of the procedure. For the procedural action to be representative of a particular outcome, the perceptual perspective must be used by the agent as a constituent in the procedural action as much as the bodily movement must be used as a constituent.

19. Some sentences can have a strong conditional or counterfactual point. 19a. Some observational, investigative procedural activities can have a

strong conditional or counterfactual point. A native has noticed that when fish eat things they can be found in the

stomach of the fish. He has also noticed that they eat different things at different times. When he catches a fish, he opens its stomach to see what it has been feeding on, so that he can use it as bait.

On one unsuccessful day's fishing he notices an approaching storm that looks like spoiling the fishing for a long time. Frustrated, he intends not to return to the fishing hole until the weather changes. He picks up his fishing gear and starts for the cave. He happens to frighten a mink eating a fish. His curiosity overtaking him, he opens up the stomach to see what the fish had eaten and takes out some grasshoppers. This is a procedural action whose projected outcome is information about the past. It also has the point of finding out what would have helped him to catch fish if he had used it as bait. He takes a moment to confirm this by throwing in some grasshoppers and watching the fish rise to take them. Miserable that he cannot use the bait that he now knows would work, the threatening storm forces him to hurry to the cave. The storm unexpectedly appears to begin to clear and he gathers his fishing gear and watchful for the reappearance of the storm he gathers grasshoppers.

20. In language one can argue that if causal factors A and B occur then effect E occurs. (Though it may be known that E may occur in the absence of A and B and by means of other causes.)

The speaker should be able to make a difference between checking on the occurrence of A and B and the non-occurrence of E versus checking on the adequacy of the causal argument itself.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

288 Proto-Language

20a. Without language an individual could have procedurally got on to the general causal ways of things, as a carpenter can know the many, many causal ways of various kinds of wood, or a master chef can develop causally general (what other kind are there?) recipes.

If the chef knows that ingredients A . . . N go to produce Z (though he knows Z can be produced some other way without A . . . N , that is, he has another recipe for Z) then when A . . . N seem to have been added and yet Z is not produced, he has two alternatives checking procedures to follow. He may check on whether A . . . N really were added and whether Z really was not produced versus checking on the adequacy of the general recipe procedure itself by seeing if the degree of heat was or was not a relevant though neglected factor.

21.There can be a first person singular use of words in a sentence. 21a. There can be a first person singular function of steps in a procedure. A young person takes a choice morsel belonging to the elder of the tribe.

The elder is angry when the elder sees it is gone and looks to each of the young people present. They know all will be punished if the guilty one does not own up. The other people become distressed and the miscreant steps forward, showing remains of the stolen food in his hand and accepts the punishment.

22. Sentence- sentence implicature. 22a. P r o c e d u r e - procedure implicature. A procedure and its projected outcome fulfilled, as in New Guinea a gift

of pigs may be actually made by one tribe to another, but the accepted procedural implicature (that the recipient tribe should at a later date itself give back at least as many pigs) is not performed.

23. Question. 23a. A quest is a procedure that is structurally directed to achieve the

projected outcome of finding out something not fully known. 24. Order. 24a. A proto-order is a procedure followed for the projected outcome of

making someone do something partly by the person's knowing that was the point of your procedure. For example, pushing someone to get them to go the rest of the way.

25. Lying. 25a. Hiding, seriously or as a joke, may be deceptive but not a lie. As

Brian McLaughlin pointed out to me in correspondence, lying requires that for X to lie to Y about p, (a) X must intentionally represent (by some action) itself to Y as believing that p, and (b) X must take itself to be thereby misrepresenting its belief to Y, and (c) X must intentionally attempt to conceal from Y the fact mentioned in (b).

X hears a wild goose calling from a meadow. X believes that the wild goose will fly to a particular pond. X is known in its tribe as a great hunter from whom the others in the tribe take direction and guidance from X's example in matters of hunting. On hearing the wild goose calls, X makes sure the other hunters can see where he is going and then starts running in a direction

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13

C. B. Martin 289

that leads to only one hunting spot, namely, the meadow. X lets the other hunters run past him toward the meadow. When they are out of sight, X changes direction, goes to the pond and kills the wild goose and eats the prized meat on the spot.

26. Promise. 26a. A Proto-promise is doing something in order to get people to think

they could depend upon you to do something else. There is an animal that is about to break into a cave with several people

in it. There is little time to spare. Only one person can go to try to kill the animal because others are needed in the cave. It is a matter of which one will go. You stand and gather your weapons and put on your leather armour. You do all of this in the sight of the others to declare that you will kill the animal and to make them depend on your going because the duration of your preparation excludes anyone else doing it. You then don't go through with it. You broke your proto-promise.

27. Homophonic and non-homophonic sentence translation. 27a. Homophonic and non-homophonic procedure translation. In watching a silent film you can recognise a procedure followed as step

by step the same procedure that you have followed. Or you see a procedure followed as step by step a different procedure from the one you have followed but you take it to be a different procedure to perform the same operation.

Conclusion

I have tried to develop a seminal model of a pattern of parallels between linguistic activity and nonqinguistic procedural activity. To a very surprising degree, racially and individually, we #ve it before we talk it. The more clearly we see this, the more natural does the easement f rom the non-linguistic to the linguistic become.

Even strikingly simple cases begin to occur to one. It would be a natural extension of the reaching procedure for it to be made and used as a pointing, because the natural reaching procedure already has the point of directionality to something not within immediate grasp. That could help to explain the universality of the pointing gesture.

Not all of the universal grammatical niceties referred to by linguists would have such tidy explanations . . . . How much and what and how of innate structures (still largely a posit of apriori gesturing) we need for the transition from our semantically sophisticated procedural activity to language gets a different look through the proto-language model, but it is too soon for answers.

Dow

nloa

ded

by [

Geo

rget

own

Uni

vers

ity]

at 1

2:52

26

Sept

embe

r 20

13