Of course, simply to formulate the question above as such already implies that biolinguistics is a kind of linguistics. And indeed, this is the idea I want to address: that biolinguistics is a kind of linguistics. What this means is that I do not see biolinguistics as a mixture of biology and linguistics or as an application of biology to the study of language (if such things could be possible at all). Rather, what I mean is that biolinguistics is the name we give to a kind of linguistics (= ‘the study of languages’) that forms a part of natural science. Or, in other words, biolinguistics belongs to the discipline that studies human languages from the viewpoint of natural science. Therefore, when I claim that biolinguistics is a kind of linguistics, I intend not to deny that biolinguistics is a biological study of language, but only to point out that the expression biological study of language should be understood not as the application of (current) biology to the study of language but as the amplification of biology so that it can study language. This means that biolinguistics should be considered an abstract layer of the biology of language, a decisive step for the integration of the study of language in natural science.
When I speak of degrees of abstraction in sciences I am actually talking about scales of complexity in the sense of Murray Gell-Mann (1994). If we consider the relationship between physics, chemistry and biology, we can say, according to this author, that physics is more fundamental (less complex) than chemistry, because the laws of chemistry are derivable from physics, “provided the additional information describing suitable chemical conditions is fed into the equations” (Gell-Mann 1994, 110-111). Moreover, those conditions are special in the sense that they do not hold throughout the Universe. Meanwhile, biology is even more complex: “[a]n enormous amount of specific additional information must be supplied, over and above the laws of physics and chemistry” (Gell-Mann 1994, 114) to characterise biological terrestrial phenomena, what allows us to conclude that “the science of biology is very much more complex than fundamental physics because so many of the regularities of terrestrial biology arise from chance events as well as from the fundamental laws” (Gell-Mann 1994, 115).
The emergence of the nervous system, of the brain, of the human brain, and of the language faculty added huge amounts of additional special information to feed into the fundamental equations of physics, and caused the subsequent emergence of disciplines that address these areas of complexity (neurobiology, psychology, linguistics, and other cognitive sciences). As Gell-Mann suggests, “the enterprise of science involves investigating those laws at all levels, while also working, from the top down and from the bottom up, to build staircases between them” (Gell-Mann 1994, 112). Biolinguistics has both missions: investigating the “additional information” of language structure and, at the same time, contributing to the building of staircases in search of unification and principled explanation.
From this point of view, it seems clear then that languages, these things that we call, for example, Russian and French, are not improper objects of study for this discipline but, on the contrary, are in fact its proper and main objects of study. Asking biolinguistics to study the human faculty of language directly while ignoring languages would be akin to asking biology to study life without studying living organisms. Obviously this would be impossible and absurd. But, curiously, the prospect does not seem as impossible and absurd to those of us in the domain of language. It could be said in principle that the biological study of language should be conducted by analysing the brain and the genes and that the study of Russian or French is a matter to be kept separate. However, this is a common mistake. You cannot study language if you do not study at least one language. To put it in other terms: if you want to study the biology of language you need more than current biology; you need a special type of linguistics: biolinguistics.
The two main traits that characterise biolinguistics as a science in relation to other types of linguistics are methodological naturalism and internalism.
The expression methodological naturalism simply implies that biolinguistics is a kind of linguistics that uses the same methodology as do the natural sciences. Of course, this methodological program does not justify in itself that we call the discipline biolinguistics. If anything, it would justify us in calling it natural linguistics or something to that effect. McGilvray (2013, 46) suggests the more appropriate (yet—in his own words—awkward) “bio-chemico-physico-compulinguistics”. In fact, one might rightly object that the bio– morpheme of our term is present because Chomskyan linguistics postulates that the object of inquiry (that is, language) is a natural object (and more accurately a biological one) and not just because it uses the same methodology as the natural sciences. It is pointless to posit a methodological naturalism for the study of something if you do not believe—or suspect—that this thing is a natural object, and it is pointless to use the form bio– if it is not assumed that the subject has biological foundations. In fact, Chomsky has repeatedly stated (see, for example, Chomsky 2000, Chomsky 2002) that language is a mental organ; he has also added that the ‘mental’ is a part of the ‘real’, just like the ‘electrical’ or ‘chemical’, so that language is just another natural object. If anyone has spoken about nativism and natural conditioning for language in our species, it is Chomsky. Nevertheless, (to my knowledge) he has never been very keen on the term biolinguistics, perhaps because his conception of science does not really grant importance to the distinction between calling something mental, neurological, chemical, magnetic, electrical or physical; these matters are purely empirical ones that depend on the historical degree of development of various disciplines, whereas the relevant point is to consider language as a natural object. This reasoning explains why Chomsky’s intellectual commitment has been always to methodological naturalism (i.e., ‘the mental’ and ‘the physical’ must be dealt with using natural science) and not to ontological naturalism (i.e., ‘the mental’ is part of ‘the physical’):
Unless offered some new notion of ‘body’ or ‘material’ or ‘physical’, we have no concept of naturalism apart from methodological naturalism. (Chomsky 2000: 143)
In strictly logical terms, ontological naturalism should precede methodological naturalism, whereas ontological dualism (the belief that the mind is not part of nature) should precede methodological dualism (the belief that natural science is appropriate for studying nature but not the mind). We could ask, then, why Chomsky starts with methodological naturalism to arrive at ontological naturalism—a shift that, if not well understood, may cause confusion and misunderstanding. The most reasonable explanation is that the more overtly logical shift (that is, the reverse: the move from ontological naturalism to methodological naturalism) is really ineffective.
It seems clear that the split between the so-called ‘two cultures’ (the sciences and the humanities) is a manifestation of methodological dualism. What is not as well known is that not every instance of methodological dualism is also an instance of ontological dualism. Of course, there is a lot of ontological dualism behind the methodological dualism in which we are immersed; many people think that issues such as language, consciousness, ethics or feelings belong not to the realm of the natural sciences but rather to the world of so-called human sciences. Nevertheless, even in the domain of modern cognitive science, it is not strange to discover that ontological naturalism (which everybody claims to support) gives rise to methodological dualism and not, as expected, to methodological naturalism.
There may be several reasons for this surprising fact, but I find that the main one is the inherent difficulty of the logical path from ontological naturalism to methodological naturalism. If we start from ontological naturalism, we must assume that the natural sciences (that is, physics, chemistry, biology, etc.) should be sufficient to explain mind and language, for example. However, it is obvious they are not. Neither physics, or chemistry, or biology as they are today can explain or predict the structure and meaning of a passive sentence. These sciences cannot even begin to adequately describe a morpheme or a phoneme, let alone the system of agreement rules in Russian. Everyone accepts that language exists somewhere or somehow in the brain, but the truth is that we are no closer to understanding a simple affirmative transitive sentence in strictly neurological terms than we are to travelling outside of the galaxy and coming back to tell of it. Thus, the more frequent conclusion is that if these things (i.e., passive sentences, phonemes, morphemes or constraints on constituent movement) are not physical, chemical or biological objects, then they either are irrelevant or belong to the realm of other non-natural, purely descriptive sciences. Here it is methodological dualism emerging from ontological naturalism—that is, a return to the two cultures.
Although it might seem surprising, many physicists, chemists and biologists believe this, and so do a significant number of linguists. However, what is implied in Chomsky’s methodological naturalism is that if any theory of language structure is empirically adequate, then that theory is already part of the body of scientific, naturalistic research on language. This stance makes sense if we recognise that we cannot prejudge what kind of physical reality language will have, and if we limit ourselves to studying it like just another natural object. This implies that the discipline at hand, although it does not work with bosons, isotopes or proteins, is a natural science. Biolinguistics is, then, a branch of this ‘natural linguistics’.
But the stance that this kind of linguistics is part of natural science is frequently rejected—not only by biologists or physicists but also by linguists and philosophers, which is even more surprising. Why are so many people reluctant to accept that linguistic theory is already a part of natural science? The first reason is, of course, that many people simply do not believe that language is a natural phenomenon; they instead see it as something purely external, of a social or cultural nature. However, even for those who accept that language is a property of the mind and brain, there is great resistance to the idea of (bio)linguistics’ naturalistic character. Here, the most plausible explanation is that the concept of natural science in itself is inadequate.
In fact, there are two main approaches to science: the empiricist conception and the rationalist one. According to the empiricist view, the goal of science is to discover the causes and nature of things, whereas according to the rationalist view, the aim of science is to translate nature into the language of mathematics. Contrary to popular belief, the task of science is not to find concepts or representations of the entities that compose reality but to construct mental realities (concepts and theories) and try to determine through experiments which ones find support in what we perceive. In this sense, biolinguistics cannot be different: It should consider how to construct formal models and theories that make the object of inquiry intelligible.
The empiricist point of view implies that reality exists outside and that the mind is able to represent it and, in a certain sense, understand it. The rationalist point of view recognises that the world is not comprehensible in itself and that the only things that we can understand are theories about the world:
“The standard that inspired the modern scientific revolution was abandoned: the goal is intelligibility of theories, not of the world” (Chomsky 2002, 68).
Starting from a methodological naturalism thus implies a rationalist conception of science; it is assumed that what is real, what exists, is not predetermined—and that, of course, it does not constitute ‘the material’ or ‘the physical’ as is assumed in narrow views of science. Physicist Steven Weinberg states that what is real is what is included in a coherent theory:
“Wave functions are real for the same reason that quarks and symmetries are—because it is useful to include them in our theories” (Weinberg 1992, apud Jenkins 2000, 32).
This idea is crucial and should confirm that I am not speaking about dualism or magic. Models by Weinberg and others, like Newton’s model in the past, are abstract and counterintuitive but adequately predict the behaviour of the observable world, that is, what matters. When we speak about quantum mechanics, it is pointless to say (as people often do) that entities are first postulated and subsequently confirmed once their ‘material correlates’ are located. Indeed, it is pointless to speak about ‘material correlates’ when we are considering Higgs boson or other particles or fields because they are postulated just to explain what is what we call ‘matter’. Physics is empirical in a deeper sense: Physics’ postulates acquire reality and existence not when their ‘material reality’ is discovered experimentally (an absurd suggestion when we consider wave functions or superstrings) but only when the theories of which they form a part adequately predict the observable world.
If we take methodological naturalism seriously in our approach to language, then we must admit that the same procedure should be applied to linguistics. If we adopt a rationalist concept of science, then it is clear that the task is not to apply natural sciences to the study of language, but instead to start from the abstract inquiry and then from there, thanks to methodological naturalism, extend the domain of natural science. Let me insist on it: The task of biolinguistics is not to reduce linguistics to biology, but to enlarge the concept of (linguistic) reality and, if necessary, rationally extend the domain of biology and natural science.
Chomsky has often pointed out that methodological dualism is a huge problem for this line of development of linguistics and other cognitive sciences. He refers (see Chomsky 2000, 112) to the double standard used to validate what is explicative in natural sciences and in human sciences. Hence, on the one hand, there are physics, chemistry or biology (sciences that are considered self-justified and that do not require, for example, that philosophers validate their results), while on the other hand there are human sciences such as linguistics or the cognitive sciences, disciplines for which additional evidence (for example, evidence of ‘psychological reality’) is required if they are to be interpreted realistically. At first sight, this may seem reasonable insofar as every natural science must necessarily be an empirical science. However, we must take into account that in this way, we are limiting the explicative power of one science in function of the limitations of another one. So, given that we cannot present any neurological or molecular evidence of a given grammatical restriction, it is supposed that we should think such a restriction unreal—or, even worse, think that it is not real yet. Nevertheless, this is pointless. It is what Dennett (1995) called greedy reductionism.
Taking naturalism seriously does not imply, as is sometimes suggested, that the brain exists and the mind does not. Again, this is greedy reductionism or materialistic naturalism, which is a simple consequence of an empiricist (and fundamentally erroneous and insufficient) conception of science. What is implied in a serious commitment to naturalism is that both the brain and the mind exist, although in different degrees of abstraction. This does not represent a return to dualism. It only means that we cannot say that, for example, a protein is more real than a phoneme. The evident difference between a protein and a phoneme is simply that the former is less complex (less abstract) than the latter, but this does not necessarily make it more real.
If we had demanded that linguistic theory be formulated in terms of psychological, neurological or biological reality, concepts such as morphemes, words, phrases, or ergative cases would simply not have been formulated. Neither French nor Navajo would be a licit object of inquiry. The result would be that the study of language would not be possible to pursue scientifically: That is, we would encounter a clear instance of dualism sprung from naturalism.
The mind, consciousness, meaning, emotions and of course languages are properties of the brain. We know this because we know that all of them disappear once the brain dies and because we have no serious reason to attribute them to anything that lacks a brain. Let us be clear on this point: Languages do not exist outside the brain. However, this does not mean either that they can be explained just by looking at the brain or that traditional natural sciences (physics, chemistry or biology) are the most appropriate sciences for approaching and explaining them. This is not just an opinion, either: It is an objective fact that neither physics, nor chemistry, nor biology can explain or predict the meaning or the structure of an utterance in any language. In light of this, there are two options: Either we say that these phenomena must be explained by another approach (that is, methodological dualism again), or we recognise that the ongoing methodological naturalistic research on this issue is already part of a (necessarily incomplete) scientific explanation.
We certainly must use natural science to study language, but we must do so amplifying the reach of natural science. The desired reduction of linguistics into biology cannot be achieved by trying to translate linguistic principles and entities into biological principles and entities; this can only occur if we amplify the reach of biology (and more concretely, those disciplines more related to the study of the brain). Let us recall now that the reduction of chemistry into physics implied a drastic change in physics, not in chemistry.
Biolinguistics is thus a kind of linguistics that belongs to the natural sciences, a kind of abstract biology. In fact, we could define biolinguistics as an abstract biology of languages. Needless to say, this level of inquiry is the beginning of the foundation of linguistic principles in biological terms, not the end. An abstract biology of language is a necessary step, but it is not enough. Biolinguistics has just started.
Anyway, if we define biolinguistics as an abstract biology of languages, then we are assuming that languages are abstract organisms. However, if the notion of languages as abstract organisms is going to make sense at all, we must now turn to the other central feature of biolinguistics: internalism.
It is well known that there are linguistic theories that reject the point of view of Chomskyan linguistics, especially functionalist and so-called cognitive linguistics. These alternative approaches also present themselves as the science of language, and occasionally, an author will indicate that his/her approach is methodologically naturalistic (e.g., Givón 2002, 2009). Nevertheless, I do not feel that they can be considered biolinguistic approaches. The main reason is that these theories are not internalist approaches.
Following Chomsky, I will define internalism as an approach that considers the object of inquiry—that is, the faculty of language (FL)—an internal property or organ of the mind. What this means is mainly that the primordial source of the structure of FL is not outside the mind and brain but inside it. This internalist conception implies, then, that the mind has its own structure. This idea, a typically rationalist one, opposes the empiricist and externalist viewpoint, according to which the mind is essentially a historical object. Fregean analytic philosophy has focused on entities that are external to the mind—propositions, referents—in such a way that the mind represents them. The mind and the brain are then like a kind of tool for storing and representing propositions. This is really a functionalist conception of the mind and brain, given that from this viewpoint, the mind and brain are mainly instruments.
To a great extent, functional and cognitive linguistics depend on a functionalist conception of the mind. Note that from this point of view, linguistic expressions and even languages are not objects of study in themselves but are instead a means of representing reality (whatever it is) or of communicating thoughts (whatever they may be). The relativist linguist Daniel Everett states this clearly:
My own theory, Ethnogrammar […] makes the case that language is a tool for communication and thought. Different components of language, e.g. recursion, binding, phrase structure, and so on, are themselves subtools. Numbers, color words, and the like are themselves cognitive tools. To say that Pirahã culture doesn’t need or desire certain cognitive tools is no more to disparage them than it is to criticize someone who doesn’t play golf for lacking a set of golf clubs. (Everett 2010: 14)
From this functionalist viewpoint languages are considered external to the mind (although represented on it). The most influential position in this regard is, of course, Saussure’s view of la langue as a social institution (a shared code), a conception which is widely extended among philosophers of language:
Si nous pouvions embrasser la somme des images verbales emmagasinées chez tous les individus, nous toucherions le lien social qui constitue la langue. C’est un trésor déposé par la pratique de la parole dans les sujets appartenant à une même communauté, un système grammatical existant virtuellement dans chaque cerveau, ou plus exactement dans les cerveaux d’un ensemble d’individus; car la langue n’est complète dans aucun, elle n’existe parfaitement que dans la masse. (Saussure 1916, 30)
From a biolinguistic point of view, a language is not something that the mind represents but is instead a property of the mind; it is not something that that the brain keeps or codifies, but rather is part of the structure of the brain. So, the language a person speaks is her language organ. If a language is a person’s language organ, we can say that a language is a person’s faculty of language. Significantly, this really implies that there is no substantial difference between language and languages—or, at least, no more than there is a distinction between life and living beings. The distinction between language and languages is artificial. It may be useful in some contexts, but we should not take it seriously. We can say informally that biology studies life, but it really studies living beings. In the same way, (bio)linguistics studies language, but what it really studies are languages. Saying that languages are concrete manifestations of language is the same as saying that living organisms are concrete manifestations of life. Both things are as apparently true as they are irrelevant.
In the same vein, the Saussurean externalist view of languages would say that internal languages (internal to each individual) are ‘manifestations’ of external languages, what would be the same as saying that tigers are ‘manifestations’ of the species of tigers. Both things are senseless from a biological point of view. What exist primarily are tigers and internal languages inside persons’ heads. The attribution of ontological preference to language as a collective (external) object over language as a (internal) human cognitive organ is another defining feature of linguistic functionalism.
I have suggested that biolinguistics is an abstract biology of language. Now, we can say more accurately that biolinguistics is the biology of abstract organs: languages.
[This post is a slightly edited version of some sections of my paper What are languages? A Biolinguistic perspective, Open Linguistics, 1: 71-95, to which I refer for complete references]