Building Models - Part Two
Inter-subjectivity and Induction
At one point in his early discussion of the relationship between word and object, Quine argues:
"Each of a party of observers glances at a tile from his own vantage point and calls it square; and each of them has, as his retinal projection of the tile, a scalene quadrilateral which is geometrically dissimilar to everyone else's. The learner of "square" has to take his chances with the rest of society, and he ends up using the word to suit. Association of "square" with just the situations in which the retinal projection is square would be simpler to learn, but the more objective usage is, by its very intersubjectivity, what we tend to be exposed to and encouraged in.
"In general, if a term is to be learned by induction from observed instances where it is applied, the instances have to resemble one another in two ways: they have to be enough alike from the learner's point of view, from occasion to occasion, to afford him a basis of similar-ity to generalize upon, and they have to be enough alike from simultaneous distinct points of view to enable the teacher and learner to share the appropriate occasions. A term restricted to squares normal to the line of sight would meet the first requirement only; a term applying to physical squares in all their scalene projections meets both."(p. 7)
A short while later, Quine adds:
"The uniformity that unites us in communication and belief is a uniformity of resultant patterns overlaying a chronic subjective diversity of connection between words and experience. Uniformity comes where it matters socially; hence rather in point of inter-subjectively conspicuous circumstances of utterance than in point of privately conspicuous ones." (p. 8)
If the "subjective diversity of connection between words and experience" were as chaotic as Quine seems to believe is the case, then, one might wonder how to satisfy either of the two conditions which Quine indicated were necessary for "a term ... to be learned by induction from observed instances where it is applied".
To be sure, the connection between a given word and concomitant experience may be subjectively chaotic. Yet, if this is the case, then, how will the individual ever come to appreciate the fact that instances in which the same term is applied "have to be enough alike from the learner's point of view from occasion to occasion, to afford him a basis of similarity to generalize upon"? Amidst chaos, where can there be "a basis of similarity" which is recognized by the individual and with which he or she can work in order to try to discern the character of the relationship between word and the various experien-tial instances in which the word is used?
On the other hand, the source of the alleged chaos to which Quine alludes could be, as he asserts, a function of the diverse ways in which different individuals subjectively connect words with experience. However, if this is the case, then, how can one maintain that the instances where the term in question is applied will be viewed as being "enough-alike from simultaneous distinct points of view to enable the teacher and learner to share the appropriate occasions"?
The phrase "chaotic subjective diversity of connection between words and experience" is a somewhat ambiguous string of terms, since one is not quite certain of the source of the presumed chaos. Nevertheless, whether the blame is laid at the feet of the individual's subjectivity or is attributed to the diversity of numerous subjective viewpoints, Quine's claim about the chaotic nature of the "subjective diversity of connection between words and experience" appears at best to be problematic, if not just incorrect.
This is not to say there could not be a general confusion and chaos in the mind of an individual concerning the connection between words and experience. After all, most of us do have trouble, at one time or another, with trying to figure out what a given word or term means in a context to which it is applied. In such circumstances, we may have nothing to show for our efforts but a lot of disjointed, incomplete and divergent connections between various words and the associated experience(s).
Moreover, none of the foregoing is meant to rule out the all too pervasive way in which we seem to miss one another in our linguistic attempts to make intelligible contact. In these circumstances, conversations tend to disintegrate into the uneasy situations, arguments or disquieting, frustrating experiences from which arise the chaos of our modern counterparts to the Biblical story of Babel.
Despite the above provisos, in order for terms or words to be learned from those experiential instances in which such words are used, then, one or more of the following will have to be the case. Either the learner on his own, or the teacher and learner together, are going to have to establish, correctly, the connection between the words used together and the character of the contexts to which they apply, as well as the specific aspect of the general experiential contexts in which they appear.
Consequently, whatever "uniformity of resultant patterns" of communication arises, this uniformity will emerge not because it overlies "a chaotic subjective diversity of connection between words and experience". Rather, this uniformity will emerge because it overlies an existential intersection of phenomenological/experiential fields of subjectivity that, in spite of their differences, have certain points of overlap of perceived/conceptualized experience.
Quine maintains "uniformity comes where it matters socially" - as if this adequately explained the moving force behind how individuals are able to communicate. Apparently, he feels that where things matter socially, this "mattering" is somehow able to impose a modulating and clarifying effect upon the "chaotic subjective diversity of connection between words and experience", generating a "uniformity of resultant patterns" of belief and communication. However, Quine seems to fail to take into consideration that his account really doesn't explain how individuals come to identify or understand what it is that matters in society.
What matters to the adults of a given society may set the thematic character of many of the experiential contexts which the young learner will encounter. Yet, this doesn't explain how the individual comes to understand the character of those themes or the meaning of the words which are applied to those contexts by the adults whom the learner encounters.
Uniformity, whether in relation to communication or belief, and irrespective of whether or not it concerns things which "matter socially", is possible because two or more individuals are able to establish corresponding subjective frameworks of meaning which connect words and experiences (or aspects thereof) in ways that reflect similar properties. These similar properties concern the manner in which the respective parties characterize certain of those experiences which are said to connect word and experience in each framework. These similar properties also involve the character of the logical or structural links which are said to connect word and experience in each framework.
Nonetheless, this process of noting similarities is not a matter of placing into opposition "inter-subjectively conspicuous circumstances of utterances" with "privately conspicuous ones" as one might conclude from Quine's manner of stating the issue. Instead, the aforementioned process is a matter of recognizing the following. Inter-subjective agreement is reached only by getting the privately conspicuous experiential circumstances of two or more individuals to reflect each other in a way which permits mutual recognizable identifying references to be determined with respect to the manner in which a certain word is to be applied to a specified range of experiences within the context of an inter-subjective exchange.
According to Quine in the previously cited quote, the "association of 'square' with just the situations in which the retinal projection of square would be simpler to learn, but the more objective usage is, by its very intersubjectivity, what we tend to be exposed to and encouraged in". The "more objective usage" of which Quine speaks refers to the following fact.
When a number of different people look at a square tile, they all perceive a "scalene quadrilateral which is geometrically dissimilar to everyone else's". This is due to the differential character of the retinal images which are formed according to the precise and unique character of the relationship of the observers' eyes with the square tile being observed. Apparently the reason why experiences involving retinal images of scalene quadrilaterals constitute the basis for a "more objective usage" of the term "square" is because, in inter-subjective contexts in which the word "square" is applied, the experience of the individuals involved is more likely to correspond with the character of a scalene quadrilateral than that of a "true" square.
One could concede Quine's point that instances of inter-subjective observance of a square tile, the general experience (with the exception of possibly a few people who were positioned just right) may more clearly approximate the character of a scalene quadrilateral. Nevertheless, the basis of the uniformity of applying the term "square" to what appears to many individuals in a given instance to be a scalene quadrilateral rests with the manner in which each of the people involved has built up over time a concept of the "experience of a square". (The idea of a 'concept of the experience of a square' is to be differentiated from just the ‘concept of a square' in which one has a closed, four-sided plane figure whose sides are equal in length and the intersection of which, with one another, gives expression to interior angles of 90° at the point of intersection - since one’s experience, depending on one’s line of sight, may yield results which are at odds with the concept of a square.)
One and the same square can be experienced, first, as a "true" square (i.e., one which appears square in the retinal projection in the sense that the character of the figure's appearance is congruent with the character of the description for which the term "square" acts as a means of specifying the focus of identifying reference). The "square", subsequently, can be seen as various kinds of scalene quadrilateral as one moves about the square and observes it from various distances and angles. All of these experiences (whether encountered sequentially in one experiential setting or intermittently across different experiential settings) form the phenomenological backdrop against which the conceptual geometry of the "experience of the square" develops.
Depending on the character of the experiences which a given individual has with squares, the "parameters of permissibility" might vary somewhat from individual to individual. The phrase "parameters of permissibility" gives expression to way in which an individual’s conceptual geometry establishes the conditions and/or criteria by means of which an individual would be willing to acknowledge a given experience as constituting a "square-experience" providing the specific experiential encounter manifests characteristics that fall within the structural confines of one's parameters of pre-understanding concerning the nature of square-experiences. Obviously, if one had never encountered a square, one's 'pre-understanding' concerning square-experiences would be couched in relative ignorance.
In any event, assuming there has been some degree of prior experience concerning ‘squares’, some people would acknowledge that, despite appearances, a given experience of a scalene quadrilateral retinal projection constituted a "square-experience". They would do so because the given experience fell within the parameters which had been set up previously (informally or formally, consciously or unconsciously) in relation to prior encounters with "true" squares.
Other people, with other "parameters of allowability" rooted in prior experiences with "true" squares, might either reject or doubt whether the same kind of an experience of a scalene quadrilateral retinal projection is an instance of a square. As a result, these people might wish to take a closer look from a less distorting perspective of angle and distance.
Differences may exist from individual to individual with respect to their individually derived conceptual frameworks of "parameters of allowability" for identifying a "square-experience". But whatever the differences may be, inter-subjective agreement is reached where their parameter frameworks coincide and reflect similar characters which can be referred to identifyingly, and recognized as such, by all individuals involved.
Therefore, inter-subjectivity, per se, doesn't underlie the more 'objective' usage of a term. The underpinnings of objective usage are a function of the way agreement can be reached in any given instance concerning the overlap of the "parameters of permissibility" which are embedded in distinct conceptual geometries that serve as the basis for a mutually acceptable context in which to establish identifying reference concerning a "square-experience".
Contrary to what Quine maintains, "a term applying to physical squares in all their scalene projections" may, or may not, satisfy the two conditions stipulated by Quine on page 7 of Word and Object (quoted earlier) with respect to what will "be learned by induction from observed instances where" the term - in this case, "square" - is applied. Whether the specified conditions are satisfied, or not, will depend on the "parameters of allowability" which have built up in the various individuals concerning what they would acknowledge as a "square-experience". In addition, one cannot contend, as Quine does, that "a term restricted to squares normal to the line of sight would meet the first requirement only". Instead, the aforementioned conditions are a matter of satisfying a learner that the experiential instances are "enough alike from the learner's point of view, from occasion to occasion, to afford him or her a basis of similarity to generalize upon".
The central factor in every individual's development of a "concept of the square experience" is the variations of retinal projection which occur in relation to objects known to be "true" squares. As a result, at the heart of every "parameter of allowability" framework is an aspect of the experiential field whose character reflects that of squares "normal to the line of sight". The rest of the framework of the "parameters of allowability" is constructed in relation to, and revolves about, the character of this central experience.
There is one final point to make with respect to the previously cited quote of Quine. Quine said in that quote: "In general, if a term is to be learned by induction from observed instances where it is applied, the instances ... have to be enough alike from the learner's point of view, from occasion to occasion, to afford him a basis of similarity to generalize upon".
Nonetheless, one might question whether the individual uses "induction from observed instances" of a term's being applied as the basis for learning the meaning of a term. One also might question whether generalization of any sort is used in the learning of a given term. In order to understand why one might have these questions, consider the following.
Let us assume that a given child is far enough along in development to have a somewhat, conscious recognition of the idea that word sounds have the potential for referential identification in a given experiential context. Each time an individual hears a certain word or term being applied the linguistic/behavioral setting in which the term appears can be characterized by the child in any number of ways.
In those in stances when the adult's verbal behavior forms an important part of what the child is attending to, the phenomenological field seems to manifest itself in two broad experiential categories: focus and horizonal background. Each of these categories has conspicuous components and peripheral components as far as consciousness of the experience is concerned.
The character of intentionality which comprises the focus aspect of the experience is directed at specific facets of the phenomenological field. This "light" of consciousness is often colored by (and its intensity modulated by) cognitive, emotional and physiological factors which impinge upon consciousness and affect the character of the "light" that is focused in any given instance (The term "light" is an analogy of sorts which is intended to convey the luminous quality of consciousness. As such, the objects which appear in consciousness or through it or by means of it are entities of which one becomes aware - much as one is apprized of the presence of, say, a chair when one casts the beam of a flashlight upon it. The analogy is intended to be suggestive rather than explanatory.)
Which facets of an individual’s phenomenology of experience will be focused on will be determined by a function that collectively structures in some fashion (according to the nature of the function) the following inputs: current needs; interests; external, horizonal contingencies; inclinations; aptitudes; as well as the moods, motivations and/or memories which shape and texture the character of focus. In addition, the function will be shaped by the object, action, process or situation which occupies center state in the phenomenological field of the given individual at a given time and which also lends a character of its own to the function that constitutes the focus aspect in question in the phenomenology of the individual's experiential field.
Whatever the character of the facet of the field to which attention is being drawn, then, in terms of the way in which an individual's consciousness is oriented within the context of the phenomenology of the experiential field, there will be aspects that are focused more centrally and there will be aspects which are focused less centrally. The less centrally focused aspects tend to merge into the ground or background features of the experiential field.
However, this phenomenological background is not just a featureless, vague dimension. It has themes and characteristics that are themselves functions of a variety of inputs, but which, for whatever reasons, play only a supporting role (at least, for the moment) in establishing the general context of the experiential setting.
In effect, the phenomenological background frames consciousness, and just as with framed pictures, although the portrait or drawing may represent the main theme to be attended to, the kind of frame used (in terms, say, of color, size, texture, pattern, etc.) can affect how one attends to the main theme. That is, the mode of framing can affect what is perceived.
The major difference between the two cases is that with a picture and its frame, the relationship is a rather fixed, static one. In the phenomenology of the experiential field, however, the character of consciousness constantly can shift as new inputs manifest themselves and as the focus of attention moves to different facets of the experiential field - what had been background, may become focus, and vice versa.
Even with the focus aspect of the experiential field, there will be features which are registered more clearly and less clearly or which will be focused more closely and less closely. These features will play off against, and interact with, the main themes of the focus of the phenomenology of a given individual's experiential field.
To suppose, as Quine does, that in the contexts in which a term is applied such instances "have to be enough alike from the learner's point of view, from occasion to occasion, to afford him or her a basis of simi larity to generalize upon" may not be sufficient. In point of fact, the learner must characterize or conceptualize the various contexts in which the term is applied.
This is true even if the contexts, themselves, in which the term is applied, should turn out to be different from one another with respect to various criterial features. Moreover, the individual may settle on any number of features as ground in a given experiential field.
Similarly, an individual may settle on any number of features as a focus under such circumstances. Yet, whatever the choice for ground and focus may be, the character of the link which ties, for the individual, a given word to a given experience will be a function of how the learner forges this link in terms of the features of the experiential field which the learner considers to be relevant to the foregoing process of thematization or characterization.
On each occasion that a specific term is applied to a context, experiential data accumulates for the child. Part of this data will be the gestures and actions of the adult(s) whose verbal behavior is being observed and which may, or may not, constitute important clues to the identification of the aspect(s) of the experiential field being referred and/or attended to.
Part of this experiential data which accumulates will be the actions and behavior of others (if any) who help form the contexts in which the term in question is being applied. Part of this data also will be the feedback, if any, that the child receives as he or she attends to different aspects of the experiential field arising out of the encountered contexts in which the term is applied, and so on.
All of these experiential encounters are characterized by the individual in one way or another and fed into that person’s conceptual geometry. This geometry is forming according to the manner in which the experiential co-ordinate points of phenomenological reference are being characterized and linked together through the dynamic interplay between ongoing experience and the already stored, thematized memories (i.e., pre-understanding) which form a part of the horizon that frames the current experiential field and which have been derived through previous phenomenological encounters of one experiential kind, or another.
Over time (sometimes gradually, sometimes quickly), the individual makes inferences concerning the character of the relationship between the applied term and the various experiential contexts to which it has been applied. The nature of these inferences may be a function of a belief whose character links word and experience according to the logic of the belief. The nature of these inferences also might be in the form of an hypothesis in which the individual is forging a tentative link between word and experience which could be confirmed, dropped or changed in char-acter in the light of subsequent experiences. Or, the character of these inferences might be in the form of a guess of some sort whose structure is influenced by one aspect or another of the experiential field which have been noticed in all of the contexts in which the term previously has been observed to be applied.
On the other hand, the nature of these inferences could be in the form of a correct deduction or "insight". This deduction could have resulted from accurately having characterized the experiential data which is relevant to the relationship of term to experiential context(s). Furthermore, out of this kind of characterization the individual might "derive" the character of the referential link that has been identified previously by one or more speakers, and which is now identified by the young learner or hearer.
While the precise nature of the "deriving" mechanism or process which makes this kind of inferential insight possible is more than a little mysterious, this mechanism does not seem to exhibit any of the characteristics normally attributed to the concept of "induction". That is, it does not seem to be so much a matter of going from known, present instances to unknown, future, or non-present instances, as much as it appears to be a matter of trying to grasp the character of the logic or structure which binds or links a variety of experiential contexts together.
This binding process seems to be a function of some epistemic dimension or theme whose character reflected the character of the intentional frameworks of the identifying reference of different speakers that stand behind a term's being applied to various experiential contexts. In other words, there is a conceptual process involved in inferential 'events" which links the focal character of a given aspect of the phenomenology of an experiential field with the character of past experiences that manifest congruent themes with the aspect(s) of the field now being attended to.
The congruency of these themes with the aspect of the phenomenology of the experiential field now being attended to allows one to make sense of a term's being applied to both the present experiential context, as well as all past remembered instances of the term's being applied to experiential contexts of specified character. In this respect, the inferential processes being used appear much more deductive, abductive, and/or insight-oriented in nature, than inductive.
Before one inductively can project into the future from present cases (or to absent cases from present cases), one has to be able to grasp the character of these cases. Such grasping usually requires one to consider the present cases in relation to past cases which are deemed to be relevant in various aspects, in order that one may have a pool of experiential data from which one might gain connecting insight into the character of the cases being attended to. Only after an experiential context has been characterized and integrated with other, similarly characterized experiential contexts is one in any position to make inferences about the nature of future instances of these contexts or about the nature of instances of those contexts which are now present.
Each experiential datum has a character. The structure of this character sets up parameters of allowability for inferences. These parameters indicate the boundaries within which one must work in order to show that a given inference could be tenably defended and plausibly shown to be linked to that data.
Inferences not capable of being demonstrated to be linkable to the character of these parameters would not be deductive in nature. They would be a product of some other process such as informed speculation, guesswork, learned beliefs, biases, etc.
Whatever the precise nature of those inferences may be (i.e., deductive or otherwise), induction seems to presuppose a determinate sense of the character of either future or absent cases. As a result, one cannot use induction to generate the character of the basis from which induction starts without becoming entangled in an infinite regress in which induction is forever presupposing itself.
Induction only makes sense when one has a basis of operations with determinate character from which to launch one's inductive projections. If this were not the case, an inductive inference would have nothing to draw on to give its projections direction, scope and definition concerning the nature of future or absent instances of the context at issue.
Consequently, contrary to what Quine seems to suppose to be the case, the learning of a term is not a function of inductive processes. Moreover, the learning of a term is not a matter of having "a basis of similarity [of experiences] to generalize upon".
Instead, the learning of a term appears to be a function of an inferential process. As indicated previously, the character of the inferential link between applications of a specific term and the experiential contexts to which it is applied is drawn from one of the alternatives that is permitted by the parameters of allowability which have been established by the individual.
These parameters have been established through his or her characterizing of the various experiential contexts surrounding the application of the term in question. As such, these characterizations form the experiential foundations on which the individual inferences concerning the character of the given term is based.
Although a term could be "learned" on the basis of a defensible or valid deduction, the inference still could be incorrect. In other words, one's understanding of a term might be based on an inference which could be linked plausibly to available data. Yet, the reference may characterize the link between work application and concomitant experiential contexts in a way which does not reflect, accurately, the character of the identifying reference intended by the individual whose verbal behavior is being observed by the child or language learner.
Under these circumstances, the learner could do one of two things. On the one hand, the learner might proceed to misuse and misinterpret the given term when it is applied in future instances. On the other hand, the individual might come to recognize that the character of his or her grasp of the term is out of kilter with the sort of feedback the individual is receiving from others, and/or out of kilter with the character of the experiential contexts to which the term is observed to be applied subsequently.
Irrespective of whether the inference(s) underlying the learning of a term is, in the above senses, plausible (or possible) but incorrect, or valid and correct, the inference(s) involved seems to be neither a function of induction nor generalization or any sort. As was discussed both in the last several paragraphs as well as in the case of the example dealing with squares and scalene quadrilaterals, identification of a given word's referential framework is done in terms of the character of the "concept of the experience" of that word's context of application. This usually consists of one or more central themes constituting the main character of the word in question.
For example, in the case of a square, this consisted in the character of a line-of-vision square which yielded a "true" square retinal projection. In addition, one must consider a variety of other secondary and tertiary dimensions of character which revolve about the main thematic axis.
For instance, to revert to the square example again, these secondary and tertiary dimensions of character would involve the diverse array of scalene quadrilaterals which appeared in retinal projections as a result of different angle and distance inputs to the perceptual experience of a "true" square. The ongoing and remembered phenomenology of these themes, taken collectively, encompasses the entire set of experiential co-ordinate points of reference which mark or characterize the hermeneutical shape of the conceptual geometry to which application of a given term corresponds.
Thus, an individual's use of this term or his or her interpretation of someone else's use of the term will be, respectively, a function of, or matched against, the character of the conceptual geometry which constitutes the term's meaning for the individual or for the other person(s) using the term. But, the foregoing does not serve as a basis for generalizing meaning to new experiential contexts or for inductively projecting meaning onto future or absent contexts. In fact, it indicates one must ensure that the character of any new context is such that an application of the given term (and its underlying conceptual geometry) would reflect, appropriately and correctly, the aspect of the new context's character to which one wished to make an identifying reference.
Conditioning and Association in the Learning of Sentences
While attempting to further elaborate upon his theory of language, Quine criticizes a more or less behavioristic model of language in the following way:
"But think how little we would be able to say if our learning of sentences were strictly limited to those two modes: (1) learning sentences as wholes by a direct conditioning of them to appropriate non-verbal stimulations, and (2) producing further sentences frum the foregoing ones by analogical substitution.... The sentences afforded by mode (1) are such that each has its particular range of admissible stimulatory occasions, independently of wider context. The sentences added by (2) are more of the same sort - learned faster thanks to (2) but no less capable of being learned in mode (1). Speech thus confined would be strikingly like bare reporting of sense data."(p.9)
Shortly after this, Quine indicates what he feels is missing in such a model:
"What more is needed in order to capitalize the riches of past experience is hinted in the remark ... that actual memories are mostly traces not of past sensation but of past conceptualization. We cannot rest with a running conceptualization of the unsullied stream of experience; what we need is a sullying of the stream. Association of sentences is wanted not just with non-verbal stimulation, but with other sentences, if we are to exploit finished conceptualizations and not just repeat them." (p.10)
He, then, proceeds to expand, in part, on what he has in mind with respect to the foregoing notion of the "association of sentences":
"... the power of a non-verbal stimulus to elicit a given sentence commonly depends on earlier associations of sentences with sentences. And in fact it is cases of this kind that best illustrate how language transcends the confines of essentially phenomenalistic reporting. Thus someone mixes the contexts of two test tubes, observes a green tint, and says ‘There was copper in it.’ Here the sentence is elicited by a non-verbal stimulus, but the stimulus depends for its efficacy upon an earlier network of associations of words with words; viz., one's learning of chemical theory. Here, as at the crude stage of (1) and (2), the sentence is elicited by a non-verbal stimulus; but here, in contrast to that crude stage, the verbal network of an articulate theory has intervened to link the stimulus with the response.
"The intervening theory is composed of sentences associated with one another in multifarious ways not easily re-constructed even in conjecture. There are so-called casual ones; but any such interconnections of sentences must finally be due to the conditioning of sentences as responses to sentences as stimuli. If some of the connections count more particularly as logical or as causal, they do so only be reference to so-called logical or causal laws which in turn are sentences within the theory. The theory as a whole ... is a fabric of sentences variously associated to one another and to non-verbal stimuli by the mechanism of conditioned response." (pp. 10-11)
Quine is quite right to maintain that the "unsullied stream of experience" which is sought after by those inclined toward phenomenalistic reporting is not enough to provide the richness of linguistic/experiential interaction that may be necessary to come to epistemological terms with the ontological objects we encounter in experience. Yet, Quine's solution for improving upon the phenomenalistic approach is to say: "What we need is a sullying of the stream" (p. 10).
His suggestion for how we are to accomplish this sullying process is to argue we should not limit ourselves just to associating sentences with non-verbal stimuli. He contends we also must begin to associate sentences with other sentences so that we might generate a theoretical network which binds a given non-verbal stimulus with a verbal response. Furthermore, according to Quine (and as quoted above): "The theory as a whole ... is a fabric of sentences variously associated to one another and to non-verbal stimuli by the mechanism of conditioned response."
As used by Quine, the terms "associated" and "mechanism of conditioned response" create a large degree of vagueness. This vagueness interferes with one's gaining insight into just what is going on in the individuals to whom such terms are applied descriptively with respect to the development or generation within individuals of the theoretical network which allegedly binds non-verbal stimulus and verbal response to the stimulus.
For instance, Quine mentions, in passing, an example concerning the mixing of the contents of two test tubes which results in the mixture assuming a green tint. The person who has mixed the test tubes says, according to Quine: "There was copper in it." Quine goes on to say: "Here the sentence is elicited by a non-verbal stimulus, but the stimulus depends for its efficacy upon an earlier network of associations of words with words; viz., one's learning of chemical theory."
In point of fact, the non-verbal stimulus may not have elicited anything. In other words, to say the non-verbal stimulus somehow acted upon the individual and forced the individual - due to its acting upon the individual in a specific way, to say precisely: "There was copper in it" - appears to be misleading.
Why not say: "It seems there is some copper in it," or, "I wonder why it turned green?" or, "It's very pretty, but what does it mean?" or, "I wonder if this indicates there was a specific kind of catalytic agent present in one of the test tubes, and the green tint is a clue as to what the identity of that agent might be?" or, "Maybe one of the test tubes had a new kind of pH indicator which reveals whether the contents of the other test tube is acidic or basic when the contents of the two tubes are mixed?" or, "This reminds me of that sweater I wanted to buy," or, "What do you think of that new instructor Mr. Green?" In short, there seems to be nothing in the fact that the mixing of the contents of the two tubes resulted in the mixture's having a green tint which would compel an individual to utter a specific, precisely worded verbal response such as the one cited by Quine.
The test tubes do constitute a stimulus. The mixing of the test tubes' contents constitutes a stimulus. The emergence of a green tint in the mixed contents constitutes a stimulus. However, none of these stimuli, taken individually or collectively, can be said to have elicited a given verbal response.
The stimuli are just that: stimuli. They merely represent information or data inputs which are received through the senses and which have to be cognitively processed.
For instance, first they must be characterized according to the nature of the stimuli. Secondly, the data must be interpreted according to current pre-understandings (if any) and assigned some sort of significance as to the meaning of the collective sequence of data-producing events.
This assigning process would be done according to whether or not the observed data fits into the conceptual geometry expressed in those pre-understandings. If the data does not fit in with one's existing conceptual geometry, one must generate some sort of proposal which is capable of taking into account the available data.
Finally, the individual has a huge choice of verbal responses he or she can make (if one chooses to make any at all) which may be directly, indirectly, metaphorically, analogically, inferentially or idiosyncratically related to the observed color change. What kind of a response the individual will make, if any, depends on the individual and his or her understanding, interests, motivations, and mood (to mention but a few), as well as on the circumstances and who else is present (e.g., what is said when a professor is looking over one's shoulder might be quite different than when a close acquaintance is standing nearby).
All of the factors (whether internal or external) which impinge on the individual's experiential field at any given time and which shape the general phenomenological character of that field and its specific focus are all stimuli. Thus, to try to maintain, as Quine is inclined to, that a given "sentence is elicited by a non-verbal stimulus" appears to be both misleading and oversimplified. To say verbal responses bear a complex and dynamic relationship to the underlying phenomenology of the experiential field out of which they emerge seems much more accurate and appropriately complex.
Quine does note, as pointed out earlier, that "the stimulus depends for its efficacy upon an earlier network of associations of words with words". In the case of the example he cites, the "network of associations of words with words" is the chemical theory that has been learned by an individual.
As such, this theory constitutes one of the major sources of the stimuli that shape the phenomenological texture of the individual's experiential field in the test tube example. However, if a stimulus' efficacy to elicit, allegedly, a certain verbal response depends "upon an earlier network of words with words", then, perhaps the stimulus really is not eliciting a certain verbal response.
Instead, one's understanding of the "network of associations of words with words" (i.e., the underlying theory) may be preset in various ways to manifest itself if certain conditions are satisfied or believed to be satisfied. In the present case, one of these conditions is this: the mixing of two test tubes results in a green tint for the combined contents. Yet, other conditions may be necessary, too - namely, how quickly the green tint appears after mixture; the temperature and pressure under which the test tubes are mixed; how pure the ingredients are which are being mixed; the size of the amounts which are mixed; and, of course, whether one knows the identity of the contents of the test tubes being mixed.
In the latter instance, if one doesn't know what is being mixed, but one is being asked, let us say, to deduce the contents of the two tubes on the basis of the observed reaction, then, one's verbal response might depend on the precise character of the "network of associations of words with words" which form the chemical theory one knows. As a result, one couldn't tenably maintain that the green-tint-stimulus necessarily elicits any response at all.
The dominating forces or factors which could be said (if anything could be said) to elicit the verbal response are a function of the perceptual/reflective/interpretive/inferential processes which generated the verbal response in question. In addition, because various people might respond differently to the same green-tint-test-tube stimulus, one cannot conclude, with any confidence, that this particular stimulus is what must have elicited the given response. More likely than not, one would have to infer that the source of the differences in verbal responses lay elsewhere, within the individual concerned.
A stimulus, in and of itself, has no capacity to elicit any response, verbal or otherwise. What happens in a given stimulus-context will depend on both the character of the being or organism which is exposed to or encounters the stimulus-context, as well as on the character of the stimulus-context itself and how that context is capable of engaging and being engaged by the given organism, as well as how that context has, in past instances, engaged and been engaged by a given organism.
Let us suppose that one knows the exact contents of the two test tubes, and all the necessary conditions are satisfied to enable the individual to identify, determinately (on the basis of his or her underlying or background "network of associations of words with words"), the reasons why the mixture of the contents of the two test tubes had a green tint to it. Even in this case, there can be no guarantee that a specific verbal response will be forthcoming or elicited.
In fact, as previously suggested, there is no guarantee the word "copper" or the word "green" or the words "test tube" or the word "mixture" will appear in any forthcoming remark concerning the observed happenings in relation to the two test tubes being mixed. One easily could come up with plausible sentences or responses that contained none of these terms.
One also could construct sentences that contained only some of the terms, but combined the terms in ways that bore no resemblance to the sentence: "There was copper in it." This is the sentence that Quine cited which would be generated by an underlying "network of associations of words with words" when the individual is presented with the appropriate eliciting stimulus.
A large part of the problem with Quine's account, at this point, is he believes the means by which words are "variously associated to one another and to non-verbal stimuli" (p. 11) is a function of conditioned response. Unfortunately, he provides no account of the nature or character of the mechanism(s) responsible for such conditioned responses or why and how the various associations of words to words or words to non-verbal stimuli come into being.
Moreover, he doesn't provide an account of the nature of the principles which forge such associations. However, he does make, from time to time, and in an approving manner, some vague allusions to the work of B.F. Skinner. Quine speaks of the "network of associations of words with words" or the theory which supposedly binds a given stimulus with a given response. Yet, he doesn't account for how a given network or theory comes to have the character it does or why words are associated with one another or with non-verbal stimuli in the ways that the network or theory stipulates.
Quine does say "there are so-called logical connections, and there are so-called causal ones" which exist in a given network of associations. Nonetheless, he goes on to claim that "any such interconnections of sentences must finally be due to the conditioning of sentences as responses to sentences as stimuli". Therefore, for Quine, all logical or causal laws are themselves merely "sentences within the theory" or "network of associations of words with words" and words with non-verbal stimuli.
Even after digesting the foregoing considerations, one still wonders why different sentences have the character they do. For example, why do some sentences come to have a logical character of a specified nature, and why do some sentences have a causal character of a specified nature? Or, why are certain words associated with non-verbal stimuli in a certain manner of specified character? Furthermore, why are various words associated with one another in different contexts of modulated character?
One also wonders how any of these sentences come into being in the first place. In other words, why do different people come up with sentences of such diverse character when they (the people) are exposed to many seemingly very similar stimulus-circumstances?
Quine's answer to all of these musings, questions and wondering lies in the idea of conditioning, but this idea is never really explicated or delineated in any clear fashion. ‘Somehow’ an individual is exposed to a context of given stimulus character. As a result, ‘somehow’ a fabric of sentences is built up over time in which words are ‘somehow’ associated with one another and ‘somehow’ are associated with non-verbal stimuli.
‘Somehow’ some of these words and sentences take on logical hues, and sometimes, ‘somehow’, they take on causal hues. However, the logical and causal hues which words and sentences take on are merely a function, ‘somehow’, of other sentences within the broader "network of associations" that constitutes the theory which intervenes between stimuli and response. In short, the idea of "conditioning" is, for Quine, largely an empty term which serves as a conceptual place holder that refers to aspects of the processes of learning and understanding which Quine can't, or doesn't bother to, pin down with any specificity.
Neither the association aspect of Quine's "network of associations of words with words", nor the network aspect of such associations, can be, strictly speaking, a matter of either words or sentences. A network is not merely a collection of random or arbitrary words and/or sentences. It is a collection of words and/or sentences connected, arranged, structured, organized or ordered in a particular manner. This ordering or structuring is done according to some given set of values, beliefs, principles, ideas, interpretations or understandings concerning the interactional character of the words and sentences involved in the network.
Moreover, an "association" is not a synonym for words and/or sentences such that what links words with words or words with non-verbal stimuli is merely more words and/or sentences. The idea of the association of words with words and so on refers to the character of the conceptual link which brings one word or sentence of a given character into juxtaposition with another word or sentence of a given character. This process of juxtapositioning generates: word/concept/word, or word/concept/sentence, or sentence/concept/sentence contexts. These contexts are shaped by the manner in which the concept(s) in question ties together given words and/or sentences, as well as by the manner in which, once tied together, the characters of the words and/or sentences modulate one another within the individual's hermeneutical framework.
Words and sentences may be used to describe, characterize, explain, identify, or refer to the nature of these networks and associations. However, the network and associations are not, in and of themselves, words or sentences. At least one cannot contend the foregoing if one wishes to avoid the pitfalls of trying to explain intelligibly what it means to say that the "associations of words with words" is a function of words, or that the network of those associations is also a function of words.
The problem here centers on the character of the function involved. This function links words with words in the form of an association, or it binds "associations of words with words" into a cohesive, identifiable network of given logical character. If the character of the function is, ultimately, a matter of words and sentences, then, in what way can one say that words and sentences link words with words to generate associations or networks?
As indicated previously, the nature of the link or function proposed by Quine is that of conditioning or the conditioned response. However, as also discussed previously, this kind of proposal is totally inadequate as it stands. As such, it would have to be developed substantially before one might even begin to consider it as a plausible account of what precisely is entailed by the terms "network" or "associations" in the pre-sent context.
According to Quine, "actual memories are mostly traces not of past sensation but of past conceptualization". One wonders what Quine means by the word "mostly". It seems to imply there are some memories, or parts thereof, which are not traces of past conceptualization. If this is the case, then, what exactly is the nature of these traces? Are they traces of past sensation? If so, is the nature of the trace similar to that of the unsullied stream of the phenomenalistic reports of sense experience? And, if not unsullied, then, what is the na-ture of the process by which sense experience is altered or contaminated? Answers to these questions would help fill out Quine's position in a clearer manner than is presently the case.
In addition, one cannot help but wonder exactly what Quine has in mind in relation to the notion of "conceptualization". Let us suppose, as Quine's position at this point seems to imply, that conceptualization is merely another way of referring to words and sentences and stands for nothing apart from words and sentences.
If the foregoing supposition is accepted, for the sake of argument, then, almost by definition a pre-linguistic child cannot conceptualize because the child has no words or sentences through which he or she somehow is able to forge networks and associations of words with words and words with non-verbal stimuli. Yet, if a pre-linguistic child cannot conceptualize, then, the child has no means, for the most part, of storing memories, which Quine has described as "traces ... of past conceptualization". But, if a pre-linguistic child cannot conceptualize, and, thereby, has a very limited means, if any, of generating memories, then, one wonders how the mechanism of conditioned response can act as a medium for the learning of words, sentences and the networking and associating of those words and sentences.
If conceptualization is a strict function of words and sentences, and if it has no autonomous dimension apart from words and sentences, then, Quine has to explain how the mechanism of conditioned response, which is to be the vehicle through which words and sentences are learned, seems to presuppose what, from a Quinean perspective, cannot exist in a pre-linguistic child.
The only way for Quine to extricate himself from this difficulty is for him to both tighten and clarify his use of the notion of "conditioned response". Furthermore, he needs to show the character of the relationship between "conceptualization" and "conditioned response".
In other words, he must explain just what it is that the former brings to the latter, thereby helping lend to a given conditioned response the character which this response has in any given case. Yet, he would have to accomplish all of this without abandoning his belief that theories, networks, associations and logic were, ultimately, a matter of words and sentences. Quite frankly, I don't see how Quine could manage to do this in any plausible fashion.
If conceptualization is not strictly a matter of function of words and sentences, then, room is left open for theoretical maneuvering with respect to developing a model of language and language acquisition. This maneuvering room would have potential ramifications for re-thinking the notions of "network", "associations" and the "mechanism of conditioned response", all of which play such important roles in Quine's approach to describing language and its relationship to the objects of the world. Such an altered conceptualization of "conceptualization" also would open up the possibility that the issue of logical connections between words and words, words and sentences, or words and/or sentences and non-verbal stimuli may not be necessarily "due to the conditioning of sentences as responses to sentences as stimuli". Instead, these logical connections may be an expression, at least in part, of extra-linguistic functions, processes and abilities (e.g., general intelligence or specialized non-linguistic cognitive abilities) which are prerequisites to, and/or run parallel with (but are distinct from), linguistic functions, processes and abilities.
|
