Note: I have embedded footnotes into the text using asterisks for formatting issues and general readability.
To watch the debate discussed, click here.
Algernon: The truth is rarely pure and never simple (2009: 54).
– Oscar Wilde, The Importance of Being Earnest, 1895
In this essay I unpack Chomsky’s conception of human nature in the context of two famous debates. The first is a 1971 televised debate with Michel Foucault in which we can observe the post-structuralist mistrust of essentialist claims. The second is Chomsky’s debate with the empiricist and behaviourist W.V.O. Quine. There we examine the philosophical underpinnings of Chomsky’s antipathy to postmodernist claims of contingent truths. I argue that this antipathy is intricately tied to Chomsky’s critique of behaviourism, though in both cases he is talking past his opponent. Suffice to say these issues are bigger than the individuals this chapter focuses on and indeed some of the debates have moved on, but the hope is that the individuals encompass enough of the core attitudes associated within their broader labels – cognitivist, post-structuralist*, empiricist, etc. – that they may act as conduits and representatives for their respective approaches.
* Where appropriate I interchangeably use ‘post-structuralism’ and ‘postmodernism’.
Chomsky’s Human Nature: A Seed Beneath the Snow
In the introduction to Chomsky’s On Anarchism, Barry Pateman writes that ‘a belief in people’s innate goodness’ is one of the ‘key components of Chomsky’s anarchism’ (2005: 10). Such a reductive statement invites a strong post-structuralist critique*. Of course, the statement is a simplistic caricature, fit for an introduction but nonetheless misleading. A more succinct description would quote Chomsky’s paraphrasing of Bakunin when he states: ‘I really would like to believe that people have an instinct for freedom’ (Edgley 2002: 51). However, the task here is to go further and understand the “innateness” principles Chomsky’s work speculates. To do so, we must delve into Universal Grammar (UG), Chomsky’s foremost scientific accomplishment.
*Not just the obvious critique of essentialism, but one which would ponder how ‘goodness’ was being conceived.
UG is Chomsky’s solution to the ‘well-delineated scientific problem’ of how a “normal” adult acquires their highly sophisticated array of abilities – in this case, language – from exposure to data which is small in quantity and degenerate in quality (Chomsky, Foucault, and Rajchman, 2006: 6). Chomsky labels this insufficient data a ‘poverty of stimulus’ (POS). We must somehow account for the gap between the POS and ‘the very highly articulated, highly systematic, profoundly organised resulting knowledge that […] [children] somehow derive from these data’ (Ibid: 3). Chomsky states: ‘there is only one possible explanation, which I have to give…for this remarkable phenomenon, namely the assumption that the individual himself contributes a good deal, an overwhelming part in fact’ (Ibid).*
*For a concise summary of the other arguments typically given, see Dąbrowska (2015: pp.4-5).
UG should therefore be understood as an innate endowment of the human mind which facilitates the selection of any of the humanly possible grammars. Chomsky’s additional speculation was ‘that in other domains of human intelligence, in other domains of human cognition and behaviour, something of the same sort must be true’ (Chomsky, Foucault, and Rajchman, 2006: 4). Here, Chomsky is positing a modular structure of the brain which houses a grammar of say music, art, or morality etc. All of the same reasons apply. One of the primary ones being the logical requirement of structure to be coherent. Chess, for example, would be meaningless without its rules and codifications. Chomsky concludes in his opening to Foucault: ‘this collection, this mass of schematisms [sic], innate organising principles, which guides our social and intellectual and individual behaviour, that’s what I mean to refer to by the concept of human nature’ (Ibid: pp. 4-5).
The contributions we make as individuals towards fleshing out these innate schema is where Chomsky gets the notion of an inherent creativity to the human condition. He therefore believes that ‘a fundamental element of human nature is the need for creative work, for creative inquiry, for free creation without the arbitrary limiting effect of coercive institutions’ (Chomsky, Foucault, and Rajchman, 2006: pp. 37-38). Of course, he is quick to declare that this is based off hope, intuition, introspection, and observation* more than anything. And so, at the end of his ‘Language and Freedom’ essay, Chomsky writes: ‘In these speculative and sketchy remarks there are gaps so vast that one might question what would remain, when metaphor and unsubstantiated guess are removed (Chomsky and Roy 2003: 406).
*An example he gives is children learning to walk. ‘Why does a kids start to walk?’ – bearing in mind they are terrible at it – ‘well they just want to do new things, that’s the way people are built […] even if they’re not efficient, or harmful […] I don’t think that ever stops.’ Chomsky thus proposes that society should be set up to fulfil these inner drives (Chomsky 2014: pp. 34-35).
A Note on Theory:
Before juxtaposing Chomsky with Foucault, it is worth noting the different vernacular at play, particularly the use of the term “theory”. For Chomsky, the word “theory” is rooted in its natural science usage. To use a theory is to find a set of principles from which one can deduce conclusions that yield empirically testable, falsifiable propositions. This traditional side of epistemology seeks to connect A to B as deductions from overarching principles. Hence, Zizek, Laclau, Foucault etc. are not using theory for Chomsky.
However, those from the epistemological camps outside of science (hermeneutics, post-structuralism) would argue that theory is not about explaining, but about understanding issues. Because social questions are so complex and intricate, we cannot fully apply an A-causes-B-epistemology, but have to think of people and their actions against changing contexts. This contextuality (norms, meaning, values) is hard to press into such a linear scheme, but developing assumptions on it would still count as theories for the post-structuralists. Drawing on Gilbert Ryle’s philosophical labour, Clifford Geertz explains the point well:
Believing, with Max Weber, that man is an animal suspended in webs of significance he himself has spun, I take culture to be those webs, and the analysis of it to be therefore not an experimental science in search of law but an interpretive one in search of meaning (1973: 5).
We therefore now turn to examine this inherited stock of knowledge called culture.
Foucault: Human Nature, Justice, and Power
As one of the most important scholars in anti-essentialist thought Foucault represents, for Wilkin, ‘the perfect foil for Chomsky’ (1999 :179). For Rajchman his discursive analyses ‘interrupt and problematise things taken for granted in our habits of thinking’ (Chomsky, Foucault, and Rajchman, 2006: 15). We can dissect his argument into three central criticisms.
The first critique is of biologism and reductionism. It is Foucault’s belief that ‘our conception and our understanding of the meaning of human nature’ is produced by the ‘dominant discursive practices of any given epoch’ (Wilkin 1999: 181). The dominant discourses (modes of understanding) are the human sciences ‘with their attempts to impose meanings of normality and pathology on the human condition’ (Ibid: 181). The ontological justifications for ideologies, especially emancipatory ones, have traditionally used a ‘representation of human nature’ as a guiding principle (Koch 2011: 24), and hence Foucault believes such declarations open the pathway to imposition via totalitarianism.*
*Foucault: ‘Doesn’t one risk defining this human nature […] in terms borrowed from our society, from our civilisation, from our culture? […] isn’t there a risk that we will be led into error? Mao Tse-Tung spoke of bourgeois human nature and proletarian human nature, and he considers that they are not the same thing’ (2006: 44).
The second critique is of homogenisation and determinism. Through the process of ‘governmentality’ Foucault observes that with the rise, in tandem, of modern political and social institutions and of the sovereign nation state, both have ‘sort to categorise, compartmentalise, and control populations by placing them into clearly defined groups’ (Wilkin 1999: 185). In Todd May’s words, governmentality is ‘the ‘practice of government’ (2006: 153). Once categorised, institutions naturally treat people accordingly, be they the criminal or the police officer – a deterministic element Foucault labels bio-power.
Finally, there is the issue of social construction. This is the idea that ‘all specifically human practices and institutions are contingent […] they are not, in any sense, to be understood as the necessary outcomes of inherent constraints imposed by human nature (Wilkin 1999: 183).
The logical corollary is that knowledge and truth is constituted by history itself. Foucault’s social ontology has a strong Nietzschean component to it, whereby ‘human identity and nature are self-creating’, transcending the modernist search for an eternal truth (Ibid: 182). Edgley concludes that ‘Foucault is not necessarily trying to take any firm position on this question of truth and its relation to social context,’ rather he is attempting ‘to redress the balance in the history ideas’ (2002: 66).
This is but an outline of Foucault’s complex writing, but it suffices for the purpose of problematising Chomsky’s argument. We might set aside the issue of human nature for now as Foucault seems to accept that it is simply shorthand to Chomsky for the innate wiring of the brain. He concedes that though the concept has been prone to plasticity in the past – used as a tool to divide and subjugate – it might be a useful organising concept for the future (Ibid). On the concept of justice, Foucault is much more wary. This is a key point of divergence. With the aforementioned critiques of Foucault formalised, let us turn to examine a brief exchange (Chomsky, Foucault, and Rajchman, 2006: 55):
Foucault: …it seems to me that the idea of justice in itself is an idea which in effect has been invented and put to work in different types of societies as an instrument of a certain political and economic power or as a weapon against that power. But it seems to me that, in any case, the notion of justice itself functions within a society of classes as a claim made by the oppressed class and as justification for it.
Chomsky: I don’t agree with that.
Chomsky offers his interlocutor a cognitivist reply similar to his previous remark on human nature. We need not quote at length, but he believes there exists ‘some sort of absolute basis […] ultimately residing in fundamental human qualities, in terms of which a “real” notion of justice is grounded’ (Ibid). It would hardly be anachronistic to say that what Chomsky had in mind was a theory of Universal Moral Grammar (UMG).*
*Personal correspondence with Professor Chomsky confirms this. However, as a neologism, UMG is meant to be metaphorical as well as technical. It does not imply that morality has a grammar in the same way a language does or that morality is similar to language in every respect (See Mikhail 2011).
For now, what should be becoming evident is the frequent discussion of related, but ultimately different, concepts. Both scholars are tying together notions which inevitably pull apart under tension. As a case in point, morality differs from justice. Where we often speak of morality is at the level of individual intent and action. Justice, Foucault’s concern, refers to the institutions whose vested judicial authority is applied regularly to coercive ends in a way which renders their power invisible because it becomes rooted in social norms, meaning, and value (recall the note on theory). Hence when Foucault replies that to objectively determine right and wrong across the population is not possible, this is not the same argument.
We cannot reconcile the two claims because Chomsky is generating his belief, as it were, from the inside out. He believes humans, their societies and their cultures are far too complex and varied to work backwards in approximating what human nature consists of – we are highly contingent on our environment. In contrast, classical theorists such as Kropotkin sought to make assumptions about human nature from observations of the vast array of cooperation (mutual aid) which takes place in the animal kingdom in spite of the Darwinian arms race to constantly out adapt a rival (see Woodcock and Avuakumovic 1993). However, Chomsky’s claim is not a post-hoc historical and cultural one rooted in neurology, but a statement about our cognitive starting point, call it S1. Foucault is concerned with the interstices between S1 and S2.*
*For the sake of argument S2 reflects some indeterminate end point. Perhaps not death per se because the post-structuralist would argue that even after death, power relations (bio-politics) are still operative (see Roberts 2009).
The question, therefore, seems to move towards how people reflexively act on their partially neurologically determined endowment. The only way to address the thoroughfare between the transition from cognitive to cultural nature* is with an epistemology. Moreover, Foucault’s philosophical heritage suggests the question should change from “what is human nature/justice?” to “what led us to accept these beliefs?” – the genealogical Nietzschean method of enquiry. If the first point of irreconcilability is the fact that there are different questions being asked which subsequently leads ‘to a difference of emphasis’ (Edgley 2000: 66), the second point is found in the reference to Nietzsche. Assuming May is correct that Foucault is an anarchist**, the Nietzsche reference illuminates a dividing line between ‘political or social anarchism and a philosophical, aesthetic, or poetic anarchism’ (Sørensen 2013: 206).
*As Wilson accurately points out, a lot of the discussion around human nature is actually about human culture (Wilson 2014: 85).
**May’s actual argument is that post-structuralism is an anarchist approach.
Chomsky’s anarchism, indeed his entire body of thought, can be viewed as a form of weak essentialism. That is to say, relationships always exist between necessary and contingent factors. We do not need to reify entities and expect them to behave in the same manner, for example comparing states to atoms or societal relations to a billiard table – strong essentialism. Take the previous example of chess; its rules and format are the necessary factors which allow us to assign it identity. Yet it is not determinate because of the myriad variations of makable moves and differences between the players – the contingent factors. Wilkin makes an elementary logical error by describing the range of outcomes generated by the necessary factors as ‘infinite’ (Wilkin 1999: 182), rather they are all-encompassing – the integers are infinite, but they do not encompass the rationals. As we will see later, Chomsky extends this form of analysis to social systems such as the media.
Wilkin states that if we do not accept the weak form of essentialism in social science, we are faced with a troubling predicament, namely:
On the one hand, it suggests that all definitions of social phenomena are thoroughly contingent and shaped solely by language and history, not by any attempt on our part to understand independently existing external phenomena. Our knowledge about objects and events in the natural and social world does not refer to the real properties of things, but constructs what those properties actually are. By extension, it also suggests that our knowledge about such objects is not separate from the objects themselves and is not fallible in any objective empirical sense (1999: pp. 194-195).
Reluctant to engage with such a position, Wilkin does not go beyond acknowledging this as untenable, indeed undesirable. In doing so he misses an opportunity to trace Chomsky’s critique back a step further, giving him a firmer foothold in the post-structuralist/post-anarchist discussion. As Koch puts it: ‘this epistemological focus decentres the understanding of politics because it suggests heteromorphous arenas for the production of truth. Language emerges in a plurality of episteme. A plurality of languages requires the decentring of politics’ (2012: 33).
Chomsky plainly disagrees and we can locate this disagreement in his famous debate with Quine. Unlike the Foucault debate, here we witness a conversation contained solely within the philosophy of language (though it has wide-ranging implications). Importantly, Quine is regarded as somewhat of a founding father to postmodernism*, but he is interesting for our purpose because he explicitly champions an ontological relativism under the banner of linguistic behaviourism**. By mapping out this lineage, we move beyond viewing Foucault as critic and Chomsky as theorist. It is clear Chomsky couches his language in mentalistic presuppositions which might not be viable, at the very least they certainly warrant inspection.
*There are those, such as Brian Leiter (1997), who believe the claim is misguided and built on poor interpretations of Derrida and Lyotard.
**In the academic literature the link between relativism and behaviourism seems to have passed without much comment. The noted exceptions are: Freeman and Locurto (1994), Benhabib (1995), and Hull (2003).
Quine and The Indeterminacy of Meaning
Quine’s thesis issues a challenge to the concept of coherent meaning in linguistic utterances and the notion that there exists consistent mental states. Thus, his critique dissolved the analytic-synthetic distinction*. The primary challenge came through Quine’s most famous book, Word and Object (1960). The book contained a probing thought experiment entitled “Radical Translation”. It works as follows. Imagine a tribe of people living in an isolated part of the world. The tribe is so remote that there is no knowledge of their language, no bilinguals or interpreters. We send a pair of linguists into the field to learn the tribe’s language, but to do so independently of each other. Naturally, they must work out the language from first principles as if on another planet. However, in principle they could come back with a dictionary and a grammar which would enable one to converse with the tribespeople and learn their language. Yet, if we looked closely at the two dictionaries they would be substantially different. Most importantly, there would not be any fact of the matter about which one was right.
*Analytic: truth by definition. Synthetic: matter of fact.
Quine elaborates by giving an example of the translation process. The linguists observe the tribespeople using the word “gavagai” whenever they see a rabbit, but language is context dependent and so we can never be sure whether “garvagi” means some platonic form of rabbit, “a set of undetached rabbit parts”, or “rabbit before the year 2100 and chicken after”. We can never verify what Quine calls the ‘intrasubjective synonymy’ of “gavagai” and “rabbit” and are thus left with the ‘indeterminacy of translation’ (Quine 2013: pp. 25-27).
Quine concludes that this challenges any notion of fixed meaning as, for him, translation is almost identical to understanding. Moreover, by extension, this means Quine is therefore challenging the very notion of essence (the intrinsic nature of something). He famously states: ‘meaning is what essence becomes when it is divorced from the object of reference and wedded to the word’ (Quine 1951: 23). To help us understand this point, Quine draws an analogy to relativity in physics: ‘querying reference in any more absolute way would be like asking absolute position, or absolute velocity, rather than position or velocity relative to a given frame of reference’ (2013: 156). The point here is that there is no such thing as absolute position or absolute velocity because space is relative as are the objects of reference. The question is meaningless. This is an important point to grasp. As Baird writes, ‘Quine has not tried to show that we cannot know what our words “really” refer to, nor has he tried to show that objects do not exist for our words to refer to; rather, he has tried to show that it makes no sense to speak in either of these two ways’ (1997: 13).
Of course, clearly this does not seem to have much relevance in our daily communications and translations. However, that is not Quine’s point, his question is how all the unlikely alternatives are recognised as unlikely and how meaning becomes obvious*. Though our daily translations are left unaffected, if Quine is correct there are repercussions for epistemology. We can spell this out a little more by contrasting the Quinean and Chomskyan accounts of what is at play in the “gavagai” example.
*This is similar to Hume’s problematisation of induction – it seems to work, but should not.
We know that language A splits into syntax (the range of available expressions) and semantics and pragmatics (the circumstances in which syntax takes place). We also know that this is but one possible systematisation available. When we take a sentence Z which is also a truth statement (for example, an axiom from Euclidean geometry such as “all right angles are equal to each other”), we know it must be upheld pending the discovery of new rules. Upon such discovery, Chomsky’s belief is that we must uphold the postulate Z and, as a result, declare that language B simply governs Z differently. Quine believes we simply do not have the facts about linguistic behaviour to do so, and so he advocates that Z was always regulated by a single rule, assertible at the time though not in light of subsequent evidence. As Horwich points out, ‘the presence of equally good alternative systemisations exemplifies indeterminism’ as oppose to underdetermination (Sankey, Ellis, and Horwich 1992: 99, my emphasis). The link here between Quine and continental post-structuralism should be clear.
The connection is most prominent when we turn to scholars such as Derrida and Butler. Derrida, rejecting language as anything more than a network of words and their relationships to each other, references Ferdinand de Saussure: ‘language has neither ideas nor sounds that existed before the linguistic system, but only conceptual and phonic differences that have issued from the system’ (Derrida 1982: 11)*. He thus endorses the Quinean thesis, as do all post-structuralists who are informed by his work on deconstruction. They view ‘all discourse as bricolage, literally tinkering or puttering around, the only activity possible because there is no centre, no point of origin and (therefore) no stable meaning’ (Barsky 2007: 240). It should be noted, that both Derrida and Butler distanced themselves from behaviourism but, as Hull notes, there is most certainly an overlap. Where Skinner and Watson spoke of physical stimuli, scholars like Derrida replace the notion with discursive stimuli. Hull writes, where ‘behaviourism prohibits the securing of meaning internally […] relativism prohibits its securing externally’ (Hull 2003: 521). She continues, stating that when combined with the writings of Quine, ‘utterances embedded in a seemingly arbitrary linguistic system are all that is left of meaning. It becomes, quite literally, meaningless to speak of the truth of a statement, either in reference to the external world or to an internal idea or feeling’ (Ibid).
*I do not have time to go into it here, but this obviously parallels Wittgenstein in the Tractatus Logico-Philosophicus when he states that the limits of his language are the limits of his world (1922). The notion that ideas which existed before the linguistic system cannot be contained within language is a claim Chomsky contests as naive and ignorant of contemporary linguistic theories on the origins of language. This is a vast literature, but Chomsky subscribes to the idea that language probably arrived through mutation in one individual. Said mutation enabled complex thought – the verbalisation of pre-existing, non-linguistic thought, which incidentally is what most thought is. This paved the way for categorisation abilities, superior planning and interpretation, and so on. Evolutionary advantages would accrue, certainly, but only after the gene had spread and dominated a small breeding group, giving rise to externalisation of the language capacity.
This is different from Pinker’s ideas about language evolving due to environmental pressures. Chomsky dismisses that theory and questions such as “what is language for?” as meaningless. Again, language enabled a different form of communication and thought to be expressed in the same way that our hands enabled us to drive a car, but it makes no sense to say our hands evolved for that purpose.
Chomsky’s replies to Quine and in turn the postmodernists (or indeed anybody who adopts the contingency statement Wilkin was reluctant to engage with earlier) are intriguing due to what could be construed as philosophical inconsistency. Let us deal with the first. His reply to Quine is rooted in an a priori commitment to cognitivism. It depends on the notion that semantic rules (the agreed-upon definition of words) can ‘become objective when we turn to non-behavioural evidence’ (Sankey, Ellis, and Horwich 1992: 96) – facile evidence for Chomsky which ‘can only conclude each response is under the control of different physical stimuli’ (Chomsky, 1959: 31). ‘Non-behavioural’ of course refers back to the UG paradigm previously established. Put simply, Quine and the behaviourists are talking about language as external verbalisation, while Chomsky is referring to internal language, sometimes called I-language (the internal monologue in our heads when we talk to ourselves). The dispute is over what counts as data. For Chomsky, sounds and symbols make language different, not the meaning behind them.
Chomsky thus posits that postulates such as, “all bachelors are unmarried men” and “all right angles are equal to each other” are contained within an I-language (itself stored in the brain)*. Under certain circumstances these meanings might be revised for pragmatic, rather than epistemological, reasons of simplification (Sankey, Ellis, and Horwich 1992: 101). We are thus able to ascertain a notion of analyticity in a way that the application of language becomes determinate. As Horwich explains, if a sentence in I-language at time T is a result of the principles of an agent’s language faculty at time T, it is analytic – true no matter what (Ibid: 103). Whether Quince’s scepticism about analyticity can be quelled and contradicted by Chomsky’s penchant for the objectivity of I-analyticity is still uncertain.
*We know this is the case with semantic primes. These are universally meaningful concepts innately understood (“I”, “you”, “good” etc.), which have to be combined to make meaningful statements (see Wierzbicka 1996).
Chomsky’s reply to the postmodernists/structuralists is interesting, however. Instead of integrating his response to Quine and saying the postmodernists have merely replaced physical stimuli with discursive stimuli, he writes:
Not that everything is wrong. What little I understand is often true: e.g., the anti-foundationalism, not only true, but truism, and for hundreds of years. That seems to me the problem: either truism, or unintelligible. To be fair, I’m exaggerating. When one peels away the polysyllabic rhetoric, there are often some good ideas, which could be stated quite simply (Barsky 2007: 254).
The anti-foundationalism comment is what stands out here. Elsewhere he writes that nobody has believed in foundationalism ‘since the 1700s’ (Chomsky 2012: 192). This is Chomsky’s neo-cartesianism shining through. Like the postmodernists, and Quine, Chomsky does not believe in absolute truths. However, his scepticism differs in this neo-cartesian point that truth need not be absolute; just because ‘some knowledge is dubiously constructed, it does not follow that the knowledge itself is false’* (Edgley 2002: 67). One might argue there exists an inchoate postmodern bent to Chomsky’s thought here which sits incongruously alongside a broadly rationalist position.
*We can even draw parallels to Nietzsche’s idea that the delineation of boundaries still has a utility function. Though for Nietzsche this does not stop knowledge and culture from being metaphysical.
This essay has undertaken the task of explicating Chomsky’s conception of human nature, which has been shown to reside, at its core, as ‘a question of defining the parameters of human cognition or human intelligence’ (Mallot 2012: 207). Since the task of the scientist is to build upon and advance past knowledge now comprehensible on a level considered hitherto mystifying by the pioneers of that field, Chomsky is perhaps overly modest. He maintains, contrary to Kuhn’s paradigm shift view, that there has only ever been one scientific revolution – the one initiated by Newton. However, his linguistic work has possibly opened the door for a more thorough understanding of human nature. Richard North-Whitehead famously remarked that all of modern philosophy is a footnote to Plato (Whitehead and Sherburne 1979: 39). Depending on opinion, the remark can be read two ways. Either Plato’s perspicacious insights laid the groundwork upon which we have advanced our understandings, or philosophy has not progressed much in the intervening millennia. How far we can advance beyond pure introspection is still unclear. Whatever the case, any and all future enquiries may well be a footnote to Chomsky.
We have also shown that the main theoretical and academic opposition to Chomsky, the post-structuralists, have a philosophical root, one amongst many, in the behaviourist school – Chomsky’s lifelong bête noire. Whilst the post-structuralists have a Quinean air about them, Foucault’s exegetical excavations in the archaeology of knowledge are stressing the construction of the subject more than anything. This is a claim highly compatible with Chomsky’s view about the importance of the environment as an indeterminate stimulus. Such ideas remain indispensable when we turn to the normative implications of Chomsky’s conception of human nature, his anarcho-syndicalist proposal.