Verified by 
Following the recent British referendum on the UK’s membership of the European Union (EU), the government has been attempting to figure out what British exit from the EU, or Brexit, actually means. While the new British Prime Minister, Theresa May, has repeatedly used the slogan—“Brexit means Brexit”, the unpalatable reality is that a divorce from the EU, by far the UK’s largest trading partner, may imperil the UK economy, and London’s status as Europe’s premier financial centre. And in turn, those who voted to leave the EU, once they see house prices slump, unemployment rise, the cost of living go up, and a spike in the cost of their annual vacations to France and Spain, may turn on the politicians who campaigned for ‘leave’ without having a plan for Brexit, or really believing that the electorate would humour them.
In light of this, there are now, perhaps, as many versions of Brexit as there are governmental ministers; moreover, each minister’s version of Brexit seems to change each time they are interviewed on the subject. Brexit is a shape-shifting chimera, as the UK government attempts to square the impossible circle of respecting the referendum result, and democracy, while attempting to avoid destroying the UK’s economy.
So what does this have to do with linguistic theory, and in particular, the proposal for Universal Grammar. Universal Grammar is an idea made famous by arguably the world’s most famous living linguist: Noam Chomsky. The proposal, informally, is as follows: all languages in the world are underpinned by a biologically prescribed blueprint for grammar; at birth each human infant is pre-programmed with neural micro-circuitry that prescribes the fundamental principles of human-like language. Not only does this biologically-prescribed, grammatical knowledge explain the broad similarities found in the grammatical systems of the world’s languages, it also, so the argument goes, enables children to acquire language, any language, in the first place.
As I have argued previously, in my 2014 book, The Language Myth, and in previous posts in this blog, supporters of the Universal Grammar hypothesis have adopted, much like Brexiters, a shape-shifting approach to their object of enquiry; it is even less clear, today, than 40 years ago, what Universal Grammar actually looks like: what are the biologically-prescribed principles that underpin all languages? And this is so, because of the lack of any unambiguous direct evidence (neuro-biological, cognitive or linguistic), in favour of a hard-wired, universal, language-specific set of knowledge structures.
Following an article critical of Universal Grammar, published recently in the Scientific American magazine, the Chomskyan blogosphere has again erupted in hostility; this time, their ire is aimed towards the authors of the article, Paul Ibbotson and Michael Tomasello. And the complaints, by the Chomskyans, run to a familiar tune; first, they complain that everyone misunderstands and misrepresents them; they splutter with indignant rage. But the Chomskyan position, like Brexit, is so malleable, that it means almost anything--and nothing. And hence, the charge that their critics misrepresent them is disingenuous, as I shall explain. Second, proponents of Universal Grammar rail against Ibbotson and Tomasello’s point that Universal Grammar cannot be falsified—that it is deeply unscientific. This is also a point I have made previously--for instance, see my post: The Shape-Shifting Malleability of Universals in UG; Universal Grammar is indeed unfalsifiable.
The Miss Martian shape-shifting complex
Miss Martian is a fictional superhero from DC comics. One of her key powers is the ability to shapeshift: Miss Martian can change her molecular make-up at will. In practical terms this means that she can choose to become super-dense or invisible; she can extend and change her limbs and body shape, adopting any human or non-human form; she can alter the chemical composition of her body, grow to an immense size or shrink into nothing. And in this way, she can repel all attacks and avoid being pinned down.
But this shapeshifting capability is exactly the modus operandi of Chomskyan linguistics, and the Universal Grammar hypothesis. In a recent blog post: Don’t believe the rumours: Universal Grammar is alive and well, PhD candidate, Dan Milway laments that the problem is: all those pesky critics of Universal Grammar just keep getting it wrong. He says: “We groan, not because we’ve been exposed for the frauds or fools that these pieces say we are, but because we are always misrepresented in them.”
But this is a familiar refrain. The truth of the matter is that Chomskyan linguistics is a study in shape-shifting malleability. Like Miss Martian, it can be anything and nothing, and sometimes, many very different things, all at the same. Indeed, any particular iteration of the theory seemingly has a shelf-life of just a few years.
As I've observed previously, back in the 1960s, the 'universals' in Universal Grammar amounted to what Chomsky dubbed formal and substantive universals. Substantive universals were grammatical categories such as lexical classes—noun, verb, adjective and adverb—and grammatical functions like subject, and object: what we might think of as the basic ‘building blocks’ of grammar. Chomsky (1965: 66; Aspects of a Theory of Syntax), suggested that languages select from a universal set of these substantive categories.
Formal universals are rules like phrase structure rules, which determine how phrases and sentences can be built up from words, and derivational rules, which guide the re-organisation of syntactic structures, allowing certain kinds of sentences to be transformed into or derived from other kinds of sentences (for example, the transformation of a declarative sentence into an interrogative sentence). But as the facts of linguistic diversity and variation have emerged, it increasingly appeared that couching universals in these terms was untenable.
By the 1980s, a revised, and more flexible approach to Universal Grammar had emerged, dubbed Principles and Parameters. Informally, the idea was that the constraints that populate our biological pre-specified language faculty, consist of grammatical principles that can be parameterised—set in different ways—for different languages. Switch the parameter one way rather than another, and you get a cascade of effects that makes a language like English look very different from, say, the indigenous Australian language Jiwarli. But in terms of the initial biological state, we all approach languages from the same starting point, prescribed by our common Universal Grammar. Summarising the state of the art, in his 1994 book, The Language Instinct, Steven Pinker summarised this view as follows:
It is safe to say that the grammatical machinery we use for English . . . is used in all the world’s languages. All languages have a vocabulary in the tens of thousands, sorted into part-of-speech categories including noun and verb. Words are organized into phrases according to the X-bar system [the system used in an earlier version of Chomsky’s theoretical architecture to represent grammatical organization] . . . The higher levels of phrase structure include auxiliaries . . . which signify tense, modality, aspect and negation. Phrases can be moved from their deep structure positions . . . by a . . . movement rule, thereby forming questions, relative clauses, passives and other widespread constructions. New word structures can be created and modified by derivational and inflectional rules. Inflectional rules primarily mark nouns for case and number, and mark verbs for tense, aspect, mood, voice, negation, and agreement with subjects and objects in number,gender and person. (Pinker, 1994: 238).
From the mid-1990s onwards, the grammatical machinery that might constitute the initial state of Universal Grammar was down-sized further, under the aegis of the so-called Minimalist programme (Chomsky 1995). The current state of the art appears to be that there is a single innate operation, termed Merge—a general purpose computation, paramaterised in different ways across languages, that enables the recursive—i.e., combinatorial potential of language(s)—such that any given language can combine syntactic units in a range of language-specific ways. In short, in the course of around 40 years, proposals as to what amounts to the grammatical information that constitutes our biological endowment—Universal Grammar—has progressively shrunk.
But, as a consequence, the changing status of what is supposed to be universal about Universal Grammar, in response to the emerging linguistic facts, has become so slippery that, that like Miss Martian, it appears to have the ability to have almost slipped into nothing; it now amounts to a barely visible target that few, save those who pray at the altar, have mystical insight to.
In his blog post, Milway claims that the most recent critical piece, in Scientific American, misunderstands and misrepresents the current Chomskyan view of recursion. Ibbotson and Tomasello are wrong because they can’t even get the claims made by Chomskyans correct. But let’s be clear. Not even Chomskyans appear to really know what they mean by recursion.
Recursion has received a great deal of attention over the last decade or so. For instance, Chomsky along with psychologist Marc Hauser, and biologist Tecumseh Fitch published an influential article in 2002, claiming that the human language faculty can be conceptualised as having two components (FLB and FLN) and that ‘‘The core property of FLN is recursion . . . it takes a finite set of elements and yields a potentially infinite array of discrete expressions’’ (Hauser et al., 2002:1571). More recently, Chomskyans have indicated that, recursion can be equated with a single computational operation: ‘‘Optimally, recursion can be reduced to Merge’’ (Berwick and Chomsky, 2011:29).
But as I have observed previously, in a co-authored paper published in Lingua, when talking about ‘recursion’ there are a number of issues that need to be clarified, which Chomskyans have, by and large, failed to do; one needs to to specify whether what is at issue is (i) the class of computable functions, (ii) recursively defined functions, (iii) recursively defined grammars, or (iv) recursive structures.
There are many formalisations of the intuitive notion of a computable function, e.g. Turing machines, m-recursive functions, Post systems, lambda calculus, combinatorial logic, and cellular automata, all of which have been proven to define the same class of functions. The class of functions characterised by these models of computation is very broad, and according to the widely accepted Church-Turing thesis, includes every computational procedure.
A recursively defined function refers to a function that is, in part, defined in terms of itself, e.g. the function for non-negative integers can be defined by recursion: (a) n! = 1 if n = 0; (b) n! = n (n 1)! if n > 0. It can also be defined without recursion: n! = n (n 1) . . .. 1, where 0! is defined to be 1.
This illustrates the point that in general, functions have various definitions. Definition by recursion is simply one type. In fact, in principle, definition by recursion is optional, and so does not describe an intrinsic property of a function. For instance, the Turing machine model of computation is an iterative one. In the more practical world of computers, at the machine code level, there are no recursively defined functions. Rather one finds loops and memory structures (stacks, etc.) that keep track of where the machine is, in the processing.
The notion of a recursively defined grammar is ambiguous. On one reading, it refers to a grammar of rewriting rules whose rule set contains some rule(s) that can be reapplied in the course of a derivation. For example, in the grammar {S -> A B, A -> a A, A -> a, B -> S B, B-> b}, the rule S -> A B can be applied more than once thanks to the rule B -> S B, which reintroduces S, and A -> a A can be successively applied to its own output. On another reading, ‘recursively defined grammar’ refers to a grammar that has some operation(s) that can be applied initially to some stock of basic elements (e.g. lexical items) but then iteratively to structures previously built by the operations.
The Chomksyan operation Merge is an example of this combinatorial approach. The upshot is that merge is not recursive in either the sense of ‘‘defined by recursion’’ or the sense of a recursive rewriting grammar. When Chomsky refers to recursion in the context of minimalism, he is tacitly referring to iteration of the merge operation. In short, it is, arguably, a category error to equate merge with recursion. So, on this score, it’s a bit rich to accuse Tomasello and Ibbotson of misrepresenting the Chomskyan view.
Communicating animals?
Milway claims in his blog that Merge, in modern Chomskyan theory, is the recursive function. He gives an example of this, essentially, iterative operation, whereby, the NP, 'my favourite book' are merged, by two iterations of Merge providing first the noun phrase ‘favourite book’, before being merged with ‘my’ providing the complex NP ‘my favourite book’. But if Merge is not in fact recursive, but rather iterative, how is this different from other symbolic capacities, such as mathematical reasoning and musical composition? Moreover, presumably concept combination is, at the very least, iterative. What then is uniquely linguistic about Chomskyan recursion? If all that is left of Universal Grammar is Merge, an iterative operation, then Universal Grammar is revealed as nothing more than an empty theoretical husk.
Moreover, how is this Chomskyan view of "recursion" different, at least in principle, from the (relatively simple) symbolic communication of animals, that, to degrees is iterative in this sense? Bees are capable of exquisitely complex dances that have a communicative function, signalling distance and direction of a food source from the hive, combining movements, locations and orientations from the sun. Campbell’s monkeys of the Ivory Coast appear to have very simple morphological capacities, with a rudimentary "evidentiality" system: they can modulate alarm calls with a suffix conveying a hearsay reading—with a "I didn’t see the predator with my own eyes, but just heard a rustle of branches that sounded like said predator" meaning. And various species of toothed whales appear to have fairly impressive iterative capabilities. For instance, some species of toothed whales have been found to have the ability to produce signature whistles--a distinctive sequence of whistles that is unique to the individual, believed to have a self-naming/identifying function. Moreover, some species may even have the seeds of a more complex ability to recursively embed structures. European starlings may be an example, and one that I discuss in The Language Myth. (Of course, in their recent book, Why Only Us, Berwick and Chomsky put plenty of clear water between animal communication and human language--although I have a few things to say about that in my review, The Language Paradox, published earlier this year in New Scientist magazine).
The Prison of Belief
In the 2015 film documentary, 'Going Clear', Alex Gibney devastating lays bare the pyscho-pathology of the Scientology sect. One of the startling messages that emerges is the way in which belief can serve as a mental prison, restricting, and even contaminating one’s world-view. This notion often seems apt when confronted with the singular inability of devout Chomskyans to 'get' why many other linguists and cognitive scientists consider their views unfalsifiable, and hence, unscientific.
In a series of blog posts, the title of which is supposed to be ironic, but is in fact prophetic: The Generative Death March, blogger Jeff Lidz argues against the criticism that Universal Grammar is unfalsifiable. In their Scientific American article, Ibbotson and Tomasello refer to the assertion, in Chomskyan linguistics, that there’s a distinction between competence—informally, a user’s mental grammar, that arises, ultimately, from the Universal Grammar—the initial state—with which a language user is born, and performance. The latter concerns, informally, language production, which might be introduce grammatical errors when we talk. The claim sometimes made by Chomskyans is that performance errors can mask competence. For instance, if a sentence is challenging for short-term memory in some way, then due to constraints on memory processing, the sentence might be judged as acceptable, even if it is actually ungrammatical. In this case, the performance system—memory—has masked the competence effect—the underlying grammatical system that would normally mark the sentence as ungrammatical. Ibbotson and Tomasello make the point that whether or not performance can be said to mask competence is unfalsifiable.
Lidz begins by mistakenly attributing to Ibbotson and Tomasello the claim that: “performance can’t mask competence”—they make no such claim--and proceeds to argue that the two can be dissociated. In fact, his argument entirely misses the point: Ibbotson and Tomasello are observing that whether or not performance can be deemed to mask competence is beside the point; the point is that competence—what Ibbotson and Tomasello refer to, in their general audience article as “pure grammar” arises from a biologically instantiated initial state; as this initial state cannot be falsified, then claiming that performance occludes the underlying competence amounts to clutching at straws.
The problem for Lidz, and others, is that they are trapped, prison-like, in their own belief systems, and cannot think beyond the confines of their mental cells. Merge/Recursion is predicated on an a priori commitment to a biological pre-specification for grammar: Universal Grammar, which, for Chomskyans, is axiomatic. You have to assume that there is a “biolinguistics”—something innately prescribed— before you can begin to posit Merge. But this principled commitment to a biological pre-specification for language, cannot be falsified—at least not on the basis of language.
But an actual “hypothesis” (e.g., Merge) cannot count as falsifiable, if it’s based upon an axiom, something which is itself impervious to counter-evidence. It doesn’t matter what anyone thinks that Merge can or cannot predict; it’s built on foundations of sand. This renders Merge, or whatever other hypotheses are proposed, vacuous. It relies on intellectual circularity.
In short, what Lidz and other Chomskyan’s don’t get is that as competence is based, ultimately, on a biological (not a linguistic) claim--that our biological endowment pre-specifies grammatical knowledge to be programmed in the brain’s micro-circuitry--then, whatever claims one might then make, for dissociations between performance and competence, are unfalsifiable. This follows as the construct of competence is based on an unfalsifiable theory-internal axiom. Moreover, even commentators who, in principle, may be sympathetic to the general thrust of the Chomskyan paradigm appear to have arrived at a similar assessment. For instance, Pinker and Jackendoff (2005) suggest that contemporary research within the paradigm proceeds on the “presumption that the Minimalist Program is ultimately going to be vindicated” (Ibid.: 222).
Reality must be able to bite, at least potentially, in the form of counter-evidence. But as the proposition—that language is biologically pre-specified—is not testable, it is not, in principle, falsifiable. And being unfalsifiable, it is, immune to counter-evidence.
