I want to sketch out here the broad outlines of where my thinking is at this point:
1. The age-old problem of Aristotle vs. Plato in this area is, in my opinion, too late to the party. The problem of do we build up kinds from instances (Aristotle) or do we rather start with innate knowledge of forms that we 'recognize' in actual instances (Plato) is so acute that we can never get started with the cognition process. This impasse suggests that something is essentially missing from the discussion--some context or self-stultifying criteria.
2. When I normally run into such an seemingly strong (and stubborn!) dichotomy, I immediately suspect (1) that the situation is more dialectical than either side represents; (2) that the context in which the opposition occurs has not be adequately examined for clues (i.e. that the abstraction of the opposing options has been excessively fine, leaving important elements of the context excluded from within the terms); (3) that the precision level of the terms are inappropriate (in either direction of more or less), (4) that the data that raised the theoretical question itself was inadequate to properly structure the options; (5) that there are metaphors that are controlling the opposition in abnormal ways; (6) that genetic and/or diachronic elements have not been taken into consideration (esp. in epistemological discussions); or (7) that the abstracting of terms process has too far removed from concrete history or occasions.
3. Increasingly, it is item number 6 that is capturing my attention in many, many questions. If we look at the hard data in infant cognition research [CS:WIK, CS:CD], we have apparently a pre-cultural 'lab'. Research into behavior and cognition in the earliest weeks/months/year or so of life provide important clues (perhaps) as to how this 'kinds' thing works. This period is important since it pre-dates most (if not all) 'cultural programming' or social construction, and therefore might yield important data for our subject. [Its impact on philosophy of mind/consciousness studies is subtantial, and I reference its important data on 'innate folk psychology' elsewhere.]
4. Let me quickly list some experimental items from this field that may bear on this subject of how we "form" kinds (pages numbers taken from CS:WIK)
(b)newborns under the age of 4 months classify colors into groups, universally! This has important implications for linguistic relativism in its various forms (e.g. Whorf, some forms of Postmodx):
"Contrary to Whorf's ideas, the newborn therefore already possesses a system of perceptual categories." (p.58)
"It seems, therefore, that the boundaries between categories are not arbitrary and that the thesis of linguistic relativism was false. A limited number of natural categories seem, on the contrary, to govern color classification. The cultural and linguistic environment can at most select certain of these categories [for lexical stock] while respecting the universal rules." (p.55)
"It seems that babies can discriminate between any two orientations with a fair degree of precision...Here again, the category of verticality is not acquired but given...It is, therefore, very tempting to say, countering the champions of linguistic relativism, that it is the processes of cognition and perception that determine the vocabulary permitting the representation of stimuli." (p.59)
"In fact, babies at birth not only possess natural categories of orientation, but certain of these even have a privileged status. Thus it has been shown that infants are capable of distinguishes a symmetrical shape from an asymmetrical one, if the symmetry is vertical." (p59)
"We can, nevertheless, say that the newborn shows a certain sensitivity to the relationships between elements. We also see that in certain cases movement is necessary for the baby to show a sensitivity to the principles of [visual] Gestalt" (p65)
"Bower's interpretation must therefore be made more explicit by saying that the presentation of a pure form, like a triangle, partially occluded by a strip of paper, does not suffice. It is the congruence of the movement of the exposed parts that allows the infant, despite the occlusion, to reconstruct the unity of the whole. " (p.87)
"..the infant's perception is guided by a conception of physical objects. Infants may group together surfaces that touch and move together because they conceive of the world as consisting of units that are internally cohesive and separately moveable." (Spelke, cited at p.87)
"Thus, for babies as for adults, the world is filled with objects. We do not have to learn to coordinate the various data from our senses in order, almost miraculously, to distill objects from them. The newborn's perceptual world is organized at birth, organized because it refers to abstract representations that are inherent and that somehow furnish schema that allow us to make connections between visual, auditory, or tactile stimuli." (p.85)
"A striking result emerged: when infants have before them a holographic image of a virtual object, they show signs of agitation if, for example, their hand passes through the apparent object. It is exactly as if they were upset by the contradiction between their visual sensation, which lead them to expect a solid object, and their tactile sensation." (p93)
"we can affirm that babies are actually capable of making correct numerical evaluations for groups comprising less than four elements. Subsequent research has yielded similar results for babies a week old." (p97)
"...implies, in children under seven months, the ability to match acoustical stimuli correctly with visual data in relation to their arithmetical properties. To explain this, one would have to postulate a sufficient level of abstraction" (p97)
"More recently it has been shown that five-month-olds can also keep track of how many objects there are in a scene, and perform what looks like mental additions and subtractions, when very small quantities are involved." (p.97)
"certain experiments show that at less than a week, babies recognize their mothers' faces." (p106) [note--these are STATIC images!]
"Be that as it may, the mechanism for the recognition of the mother's face seems to be well established at an early age, while the orientation toward stimulations resembling faces is in place at birth. It is only after two months that another mechanism appears which permits the creation of an album of familiar faces." (p106-107)
(r) Babies can tell the difference between the native tongue of their mother and a foreign language at 4 days old!
"It is undoubtedly this attraction to human speech that allows babies to disregard noises that do not correspond to a voice and, more precisely, to recognize the intonation and rhythms of their native tongue. Mother's voice would therefore serve as the model from which to determine the rules of a language." (p152)
6. The above data from infant cognition might suggest the following tenets:
8. The cultural protocols cause development of technical kinds (as opposed to natural kinds). A technical kind is a non-basal genus, such as a grouping of pink and green colors into a category called 'grink'. This could be a frequent reference within a culture, for example, and be something everybody discussed. As such, however, these technical kinds have the essential character of definitional terms or specialist jargon. They ARE constructed by cultural forces and mediated by family and other institutions to kids, and can be seriously value-laden.
9. Operational needs might be represented by refinement of the natural kinds for functional requirements related more to basic living than to symbolic living. For example, refinement of the category of 'snow' into the twelve sub-categories of 'snow' for Eskimos would be based on functional requirements--not 'fashion' or 'cultural values'. [This is not to say that some categories could not also take on metaphorical functions (pure snow vs. 'New York snow' ?), or symbolic functions. But in these cases, the operational definition is the basis for the cultural ones.]
10. The more specialized these technical kinds are, the farther away from natural kinds they can be. In mathematics, for example, imaginary numbers and non-Euclidean geometries are quite remote from natural kinds. This, of course, does NOT limit their usefulness, because they seem to function as models with which to 'model' a particular aspect of some subset of reality. Imaginary numbers are a highly 'constructed' reality (in a loose sense of reality, of course!), but are of great benefit in building 'real' composite objects like bridges and buildings and jets.
11. The above discussion focuses on physical/personal kinds, but do babies have value kinds of some sort at birth too?
12. I pointed out in the piece on consciousness ; that babies apparently have an innate knowledge of subjective mental states of other persons in infancy, and have a fully developed folk psychology by age three. We know also that from birth babies will purse their lips and move their heads in the direction of a human voice, apparently expecting feeding. When you couple this with the observations by early childhood specialists that a child forms--within 6 weeks-- the impression of the universe as a warm, need-meeting place or as a cold, unresponsive world, then you can see the rudiments of a value-system.
13. This values-grid would at least have axes of responsiveness, benevolence/treachery, gentleness/harshness (based on tone), and perhaps integrity (correlation of imputed subjective mental intention and overt action). This would make 'natural kinds' of the values of personal interaction and loyalty, goodness/kindness, respect, gentleness-warmth-caring, and honesty.
13b. The problem with this, at first blush, is that these 'values' appear self-centered, focused on the infant itself. We know that small children can be extremely possessive ("mine!"), yet are also known to spontaneously share toys or food with other toddlers. But a self-oriented value grid is not inimical to altruism, as would appear from the various versions of the "Golden Rule" in most classical religions. The 'do unto others as you would have them do unto you" REQUIRES at least an understanding of behavior would be useful/pleasing/productive/supportive of the self. Indeed, even the words of Jesus in the Second Commandment--"Love your neighbor as yourself"--links the two value-sets at a basic level.
13c. This would argue that the infant's self-oriented (or at least defined by reference to the self) value set MUST BE in place BEFORE a community/social ethic can be realized. I must know how I want others to treat me, BEFORE I can understand what behavior toward others would be "valuable" to another as well.
13d. It is interesting to notice that all that is necessary to generate the other-ethic from the self-ethic is a noun-substitution (or possibly only a negation) in the language! I need only to be able to permute "responsiveness to me is good" into "responsiveness to another is good" (or even "responsiveness to non-me is good"...
13e. How this permutation is made is not clear to me--but the fact that children seem to come up with "it's not FAIR!" so quickly argues that the concept is easily and/or very early learned. It seems that this concept only begins to appear after acculturation begins, but without some 'natural' basis, I am not sure how this concept of 'fairness' could be learned so easily. (The field of Social Cognition in children/infants is extremely complex, and changing rapidly due to the undermining of the language-based methods of Piaget, in favor of more behavior-observation methods.) We DO know that children before 4 years of age can predict the emotional outcomes (e.g. happiness, sadness, crying) of participants in imaginary scenarios. This would argue that the substrate needed to at least build an others-oriented ethic is present early.
13f. This 'leap' into the self-oriented 'self' of another brings us to the concept of empathy. We DO know that young kids reason empathetically early, and infer the subjective mental states of others on the basis of their knowledge of how their OWN subjective mental states work. But empathy is connected to the cognitive concept of 'social attachment'; and both seem to find a substrate in the innate faculty of imitation.
13g. Imitation is a defining trait among newborns. The data that supports innate facial maps and body maps is considerable (CS:WIK:100ff). The instinctive nature of a newborn to imitate facial movements of persons successfully points clearly to some kind of other-centered behavior. It is going to far to suggest that this basic feature is enough upon which to base an altruistic substrate (!), but I still think that it is suggestive of being able to put oneself in another's position--a basic requirement in empathy and moral reasoning. This may be the natural schema that allows us to make the linguistic tranformation to a community-oriented ethic.
14. As with physical kinds, so to natural values-kinds might be used to construct cultural "technical kinds" such as "2nd degree murder", or "white lies," or altruism, or 'team spirit'. To the extent a technical kind mapped onto a combination of natural values, to that extent it could easily be 'disguised' or accepted AS a natural good, when in fact it may be a set of goods hierarchically arranged.
15. If the above argument can be maintained and developed (and I think it can) then any assertions that 'all values' are culturally constructed/defined is simply a gross overgeneralization. Many (but not necessarily ALL) technical values MIGHT be constructed, but natural ones wouldn't, and even technical might be sub-divided into those (a) operational and those (b) protocol-ed.
16. What is interesting about this scenario is that conflicting values will emerge in the technical kinds, in some cases causing dissonance within the individual. If I construct some 'will to power' value at the technical-cultural level, which contradicts some 'natural' ethic toward warmth or gentleness, then I can expect some type of dissonance to manifest itself later (as we often see in such secondary groups).
17. The postmodern thesis of linguistic/categorical relativism seems
to falter in light of infant cognition studies, and I suspect the 'all
values are socially constructed' position always fails thereby.
The Beauty of Dissonance: A word about cultural 'incompatibility' or radical perspectival incompatibility...
18. Every so often I hear someone argue that categories/kinds are developed or constructed by culture alone, and the evidence advanced is often lexical in nature. For example, anthropologists will cite some tribe somewhere that doesn't have a word for 'forearm' or the Eskimos, that have 12 different words for snow (as opposed to Americans' one or two). They sometime argue that this is evidence that the world 'out there' , the world of external objects is culturally defined and/or socially constructed.
19. I have argued above that the 'hard data' of infant cognition studies argues strongly against this position, but here I want to demonstrate that even these differences are inconsequential.
20. Consider the first time I (with my lexically-challenged heritage) encounter someone with different lexical stock than I, but whom still speaks the type of English I do. I will make some sentence, which words which are phonetically in my partner's vocabulary, but my sentence will seem 'odd' to her. No real closure will occur for her, and she will hold the matter open (as I continue talking) until either the reference makes sense, or until she decides that he words are too far apart for common consensus. In the first case, she will go back and resolve the unsolved references, and 'fast forward' the semantic payload to the place I am in the discussion at THAT point. In the last case, she will likely stop me and ask "what do you mean by 'glory'?", and will admit to not being sure of my meaning.
21. What happened here? Something in the discourse/dialogue did not make sense--a dissonance emerged that she could not reconcile with my obvious intent to elucidate (not confuse). She feeds back to me that 'glory' for her has never had the content of 'a knock-down argument'. Whereupon I will either seek to amend the term, define the term, or take a different tactic. In ANY EVENT, the lack of closure/harmony will 'tip us off' that category match did NOT occur, and we have definite means to resolve the issue.
22. I can, for example, set out a calculus description and element definitions. Or I can try a series of paraphrases and lexical equivalents. I can try ostensive tactics, like telling stories and 'pointing' to aspects of the story. We can arrive at a 'working and temporary' vocabulary long enough to continue the conversation meaningfully.
23. Now, it should be obvious that this is how we acquire much of our vocabulary in real life anyway! Teachers, books, friends all use words or expressions we do not understand, but we glean a 'starter meaning' from the context. We may even ask for definitions. But in the case we are discussing here--in which we THINK we know what the word means for the other speaker--is slightly different. It is different that we must first DETECT the discontinuity in meaning, and THEN we can apply standard learning/acquisition methods.
24. It is this ability to (or natural instinct to) detect lexical and/or semantic incongruities that virtually assures us of the ability to transcend (or at least 'transgress'!) one's socio-linguistic background. I cannot be captive to my semantic competency if I (1) adapt someone else's language-system temporarily and adequately, by which to create temporary ad-hoc lexical stock; or if I (2) can detect lexical dissonance via some mechanism of semantic non-closure.
25. Needless to say, this does not work across foreign languages generally, although ostensive definitions can always be made of the object-world and to a lessor extent, the personal-world. The problem is that there must be the basic lexical stock to represent the natural kinds (from infant cognition) FROM WHICH to construct ersatz languages of a temporary sort. Fortunately, we have warrant to believe in the pre-construction nature of the categories of the human mind we described above.
26. A position that says that one cannot step out of one's socially-constructed position INTO another's socially-constructed position (1) runs counter to the examples of real-life cross-cultural learning, cultural blending, and ultimately (2) cannot be maintained with integrity.
27. I have already pointed out that language-modification via learning, via temporary ad-hoc constructions, via diachronic change demonstrates that two linguistic positions can interact, implying the ability to interact around SOME 'common semantic ground' (although discovery of that, or construction of that common ground is generally necessary). The simple reality of language translation--at a functional, literary, and technical level!--shows that people DO step out of their positions into some 'target' language or culture.
28. But even more striking is the second point--that the position of radical discontinuity between language-worlds cannot be maintained consistently or with integrity. The logic of my argument is simple: without being able to step outside of one's position to understand the position of another, one cannot even assert that the other position is DIFFERENT! In other words, to know that non-A is different from A, you must FIRST be able to know that non-A IS non-A. To know that a position is NOT the same as/equivalent to your own, you must first be able to understand its meaning and content! The very thing that is allegedly impossible to do--to have a 'universal' or 'meta-language' with which to discuss the differences between the positions--is absolutely necessary to do, in order to know that the two positions are indeed non-identical.
29. This is a strong argument, actually, and it applies to ALL versions of radical perspectivalism, paradigmatic tyranny, various types of fideism, all systems that assert that the "believer" and "unbeliever" have nothing in common (epistemically). You simply cannot talk about the content of a different position as being different, without being able to understand it adequately.
30. This, of course, is related to self-reflexive arguments. You must have a meta-language-game to discuss language-games, and you must have a socially-invariant discourse space in which to identify, discuss, and distinguish socially-constructed discourse spaces. You simply cannot avoid this transcendental requirement.
31. But this does NOT require an actual 'external' language, outside of the ordinary language space. We make meta-references all the time from within ordinary discourse. In fact, this is a very early skill in children [CS:CD:216ff], to use the recursiveness of language to refer to objects of thought. The "I thought you meant X" or "I know I am not making much sense here" or "When I said X, I was really trying to say X-paraphrase." We don't generally have to construct a full meta-langauge for this (we may have to in order to avoid language 'going on holiday'), because there seems to be enough transcendence in language already. (Indeed, philosophers and linguists alike do this everyday!)
32. But this ability of language to refer to itself raises two deep motifs that plague us--the motifs of linguistic transcendence and the issue of freedom.
On linguistic transcendence...
33. The postmodern/deconstructionist argues that since language is socially-conditioned and culturally-transmitted, then it can contain no 'absolutes'--it, and all it conveys, (and in many cases, all it refers to) is purely 'relative'. There are only models, metaphors, and perspectives--all floaters; none anchored in some external or given reality. The 'modern' on the other hand desires the early, 'unenlightened' Wittgenstein--in search of calculi. He hopes for the precision of a God, and correspondence as 'without remainder' as identity.
34. But language is inherently and internally transcendent. It transcends (in the sense of referential distance, epistemic distance) all that it refers to --including itself. Word, object, and rules of referring FORM a bubble, form a context, form a universe. The poles of the relationship--the phrase and the referent--both are part of the context and are conditioned by the context. The relationship itself creates the closed character of the linguistic event-universe. The word is 'distanced from' or transcends the object, JUST AS the object transcends the word. But this transcendence is still WITHIN the context. It is a relative transcendence, and acts as a 'working absolute' WITHIN the bubble. Hence, philosophers through the ages have held to the almost god-like quality or power of linguistic reference. It was considered pure and unchanging and outside the flux of Heraclitus. It was the absolute.
35. But then the postmodernist decides that language itself is subject to and embedded in the flux. It too is changing, impure, relative...and for them the 'transcendence' moves to the principle of flux...It is only the fact that "knowledge is socially-constructed" that is itself not socially constructed! One can immediately see that this is self-refutational. They still have absolutes--things that transcend--and in this case it is the 0-dimensional character of history/culture.
36. What do I mean by this? A 3D world has distance within it--a point can 'transcend' another point...the same is true in 2D, and even in 1D (a number line)...but in a world in which NOTHING can transcend, one loses all differentiation. One is left with a single, undifferentiated point--a world in which maya reigns. Yes, it LOOKS LIKE language can transcend--indeed, I the postmodex is using language to describe language's relationship to reality--but that is an illusion. The postmodex has (objectively) noticed that someone in the past has used language NON-objectively (e.g. as a means of marginalizing). But to know that the 'someone in the past' was abusing language requires that the postmodex KNOW what was the 'real story' . The postmodex must be able to 'transcend' the limitations of historical discourse, narrative, rhetoric, and polemic to get to some 'absolute' or 'transcendent' epistemic position BY WHICH TO JUDGE the allegedly power-oriented use of language by some elite. How the postmodex can assume such objectivity on their part escapes me (unless it is purely a rhetorical or dramatic device to 'wake us up' to some of our more 'perlocutionary' forceful speech acts.).
37. I personally have very little sympathy with any philosophical position that instructs the world to "do as I SAY--not as I DO"...we have enough of such types in the religious community...
38. The very structure of language as 'referring' (one major function of language--irrespective of one's views of the other functions of language) allows for transcendence--for judging some linguistic formulation as being (1) "more representative of reality" than a competing statement; as being (2) "more culturally conditioned" than a competing statement; and as being (3) "more self-critical of one's power motives" than a competing statement.
39. Indeed, it is only this transcendence that allows the postmodex to discuss language's nature, acquisition, limitations, abuses.
40. Now, this transcendence is STILL inside the bubble, and is still socially-conditioned...but in that moment in which it transcends the object it is pragmatically and operationally absolute. Later, when we turn language back on that statement, and turn that once-absolute statement into an 'object of referring', THEN it will not be transcendent and absolute, and it will show its conditional character...But when it was on center stage, it taught us truth (conditioned but vivid, context-ed but incisive).
41. Consider the following self-refutational statements:
a. "The vocabulary of this statement was not learned through culture by the author."42. If the Modern was guilty of deifying language and idealizing its abilities in areas of precision, neutrality, and scope, the Postmodex is guilty of deifying culture (and the lust for power) and idealizing its abilities in areas of control, consistency, and ubiquity. Culture is simply neither mere tool nor sovereign god...
b. "The meaning of this sentence is ABSOLUTELY indeterminate."
c. "The syntax of this statement is socially-invariant"
d. "The semantic effect of placing the logical operator 'negation' on a precise and true proposition is unpredictable for future situations/conditions/worlds."