Beyond Human Nature by Jesse Prinz
Critique by Michael Lopresto, 16 Jan 2013.
In Beyond Human Nature, Jesse Prinz (2012), a staunch empiricist, sets himself the task of setting the record straight against the now seemingly dominant rationalist or nativist view in psychology and cognitive science.
The modern nature-nurture debate has its roots in the 17th and 18th centuries, where the rationalists of Continental Europe, such as Descartes, Leibniz and Spinoza were countered by the British Empiricists Locke, Berkeley and Hume, and later Mill. The rationalists held that our knowledge was entirely innate: that humans are born with structures in the mind that furnish us with all the concepts we need to understand the world. The empiricists conversely held that the human mind begins as tabula rasa, and that we gain knowledge of the world purely through experience.
Immanuel Kant attempted a formidable adjudication of this debate, trying to establish a middle ground position: humans have some innate structures in the mind, which furnish us with concepts we would never be able to acquire through experience, such as time, space and causation. With these innate structures, we are then able to make sense of the “raw data” that we receive through our senses.
Some found Kant’s analysis compelling, but others such as Mill were still motivated to argue for an empiricist position in light of Kant’s innovations. Mill even arguing that such things as our knowledge of numbers, ground that was traditionally thought to be won by rationalists, could be better accounted for by empiricism.
In the first half of the 20th century, a strong version of empiricism became dominant, due to the influence of behaviourism. Behaviourism held that the human mind was born a tabula rasa, and that all human behaviour could be exhaustively explained without any reference to internal structures. I doubt that any right-thinking behaviourist thought that the skull was full of sawdust. They would have known that there is neural machinery in the head. It was just that reference to neural machinery was about as explanatory relevant to behaviour as reference to the quarks and gluons that comprise the human body.
On the philosophical side of the debate, philosophical behaviourists such as Gilbert Ryle and Ludwig Wittgenstein argued that the mind is non-causal – an idea quite foreign to us today. Wittgenstein in particular, however, did give powerful and extremely influence critiques of the nature of meaning. At the time Wittgenstein was writing, phenomenology was influential throughout philosophy. The phenomenologists held that the human mind has intentionality because it has a phenomenological “picture” of the thing it is representing. So, as I’m sitting in café and looking at the menu drawn up on the whiteboard, according to the phenomenologists I represent the writing on the whiteboard because I have a phenomenological picture of it. “How does the fact that the writing on the whiteboard has gone from the surface of the whiteboard and through my skin and skull help at all?” Wittgenstein may have asked. Wittgenstein then went on to give an extremely influential account that meaning is nothing over and above use – that meaning is coordinated behaviour, and that positing structures internal to human minds is irrelevant in accounting for the existence of intentionality.
By the 1950s, not only was psychological nativism out of fashion, but also the idea that what was in the head was relevant to our psychology. The pendulum had swung in perhaps the furthest empiricist direction that it ever had. In 1959, the pendulum began to swing in the opposite direction, and perhaps to an equally extreme extent. In 1959, Noam Chomsky, then a young linguist, published a devastating critique of B. F. Skinner’s Verbal Behaviour, from which behaviourism never really recovered. Behaviourism has made somewhat of a comeback in recent years, with those denying the existence of internal representations and even those postulating that consciousness is behaviour. But this is a somewhat more sophisticated behaviourism, however erroneous.
Chomsky taught us that it was necessary to postulate internal structures to explain verbal behaviour. Chomsky then made the further claim that our ability to learn language was innate, which became the dominant view in linguistics and psychology. To test this claim, Chomsky postulated that there would a “universal grammar” that we could discover, and which would then verify nativist view. After decades of research no such universal grammar has been found, and peoples have been discovered to have radically different linguistic practices from what we’d expect, but still some nativists are optimistic.
In the 1970s and 80s, Jerry Fodor put forward influential nativist theories of the mind. Fodor held that all of our concepts are innate: we’re born with all the concepts that humans could possibly have, and only some get used when triggered by experience. So, on Fodor’s view, cavemen (and cavewomen) had the concept in their head of stealth bomber, but never used it. This view is extreme, but Fodor was merely taking the theory of mind that he and others subscribed to to its logical conclusion. Fodor is a proponent of the classical computational theory of mind, which says that thought is produced by the manipulation of atomic symbols (concepts) according to syntactic rules.
Fodor was not only a nativist about concept acquisition but also about learning. Fodor held that learning was possible through hypothesis testing, and that the hypotheses that we test and either confirm or reject are concepts that we already have. If not, learning is just a random process, according to Fodor, which is clearly false.
Fodor also established the theory that grounds much contemporary nativism: modularity. Modules are innate and specialised mechanisms in the brain. A very moderate version of this thesis has been somewhat vindicated by neuroscience. Like the phrenology of the 19th century, there are areas of the brain dedicated to very specific tasks. There are areas of the brain that are purely dedicated to the use of words that apply to professionals, like “doctor” and “nurse”. There are areas of the brain that are dedicated to the regulation of emotion. Much of the brain is dedicated to vision, and minute damage to different areas of this area can cause radically different dysfunctions to vision, such as colour blindness and vision of motion.
Contemporary theorists more partial to nativism have argued that evolution has endowed us with modules that relate to anything from language acquisition to moral reasoning. Just off the top of my head, here are the things that I’ve heard claimed to be native, and owing its existence to modules in the brain:
- Language acquisition
- Concept acquisition
- Moral reasoning and moral concepts
- Theory of mind
- Face recognition
Unlike most theorists, who find some of these postulates more plausible than others, Prinz will have none of this. I’ll give a critique of Prinz’s argument shortly, after I finish this introductory exposition.
As I mentioned, most contemporary theorists are partial to language acquisition being innate. And most of those will then probably be partial to theory or mind and emotions. Less would be partial to a narrow form of rationality (instrumental rationality), and minimal concepts, such as causation. Still less for moral reasoning and religion, perhaps citing some universal minimal morality and some supposed universal desire to worship. (I’m optimistic that the desire to worship is purely learned, and that in my lifetime hardly anyone will feel the desire to worship. I suppose I should also acknowledge that just because something is universal, it is not at all obvious that it is then innate and pointing to a specialised module.)
I have to admit, before coming across Jesse Prinz’s contribution to the nature-nurture debate, I thought that the debate was largely put to rest. After the publication of Matt Ridley’s Nature Via Nurture about 10 years ago, I thought it was a matter of fleshing out the details in a middle-ground position. Prinz gives compelling reasons for thinking that a middle-ground position does not do justice to the debate, and that nurturist explanations of human psychology are far more informative and useful than nativist ones.
Before reading BHN, there were two cognitive faculties that I thought were pretty well native: language and theory of mind. When it came to language, I was part of the Chomskyan mainstream. When it came to theory of mind, I assumed that those with autism had native differences that impaired their innate theory of mind. In Chapter 4, subtitled “What Babies Know”, Prinz gives a compelling empiricist explanation for the impaired mind reading and mental state attribution abilities that those with autism have. Prinz says that the problem with the view that those with autism natively have an impaired theory of mind is that the view assumes that they have no other impairments (pp. 102/3). This is not the case. Particularly, those with autism are hypersensitive to sensations and also have trouble integrating information. Human beings are intense stimuli, and those with autism develop an aversion from a very early age. They also have trouble integrating all the information about other peoples’ facial and behavioural cues. Hence, early in life, those with autism never learn a theory of mind like children without autism do.
I find this explanation very compelling. However, there are still things that I think may point to nativist aspect of theory of mind. Why is it that theory of mind is extremely anthropomorphic, and that we project human psychology and even human faces onto simple animations and photos of plain circles and triangles, and even leaves and branches on a tree? Why is it that those with damaged amygdales lose this ability, that normal people have? I think Prinz would reject any nativist explanations for such phenomena, but I doubt that I’d be willing to go all the way with him.
One of the clear impediments to making progress in the nature-nurture debate, it seems to me, is actually setting up the debate in a fair and useful manner. Perhaps this hasn’t been done in some discussions, as in Steven Pinker’s The Blank Slate. Pinker takes himself to be refuting the idea that the human mind is born a blank slate, but no one today really holds this idea. Prinz is careful to emphasise that there is no one who holds the blank slate view, and conversely, that no one holds the biological determinist, “genes are destiny” view.
Also, it seems to me that “nature” and “nurture” are hard to define in the first place. I can’t come up with a definition of what it is for some to be “nature” or “innate” or “native”, because anything that would be even uncontroversially native, like vision or eye colour, needs nurture to work properly. And conversely, things that are uncontroversially nurture, like reading, are only possible because native structures that allow for vision and bodily nourishment. But perhaps such definitions and “conceptual analyses” are completely unnecessary.
Prinz avoids such definitions, and exposits the nature-nurture debate in a way that I think is very useful and interesting. Prinz sees the debate as largely between those who think that the mind comes furnished with a wide range of domain-specific mechanisms (nativists) and those who think that the mind comes with domain-general mechanisms ready to be shaped by experience (nurturists). So, how should we understand these terms?
We can say that domain-specific mechanisms are adaptations that have evolved in hominins to solve specific problems of survival and reproduction. Since our ancestors lived in rich social environments, and were evolutionarily successful because of their social and cooperative natures, it is natural to suppose that any domain-specific mechanisms that may have evolved would apply to the domain of social interaction and cooperation. From this perspective, it seems to natural to suppose that language acquisition, theory of mind and face recognition are all psychological traits that are native to hominins because of domain-specific mechanisms that have been evolutionary useful.
Some philosophers and psychologists have made greater claims to explanatory power, such as Marc Hauser and Jonathan Haidt, who’ve asserted that the same could be said for morality. For cooperation and pro-social behaviours to exist, hominins need to follow certain norms conducive to such behaviours. Hominins need to have in-built desires to not free-ride off the hard work of others, to be fair and reciprocal, and so on. Haidt has also claimed that other aspects of morality can be accounted for by an evolutionary explanation, such as our “belief” that incest is wrong (perhaps not so much a belief, but disgust).
On the other hand, it is possible that the hominin mind comes furnished mainly with domain-general mechanisms – the hypothesis that Prinz forcefully argues for. On this view, the mind comes furnished not with modules that are hard-wired for language acquisition, face recognition theory of mind, and so on. Instead, the mind comes furnished with mechanisms that can be shaped for different purposes, as the experience of the individual requires. The difference between the minds of hominins and their close relatives is minute and incremental, allowing for greater flexibility and plasticity.
As Prinz is keen to point out, the empiricist view is far more economical. We should only believe in domain-specific mechanisms if we are unable to explain the existence of certain psychological traits with domain-general mechanisms.
I think we could easily explain the existence of morality in humans without appealing to some “morality-module” in the brain. As Colin McGinn and Peter Singer have argued (shortly after Dawkins’ seminal publication of The Selfish Gene in 1976), even if our genes cause us to act “selfishly”, humans have adapted the faculty of rationality, and as Kant has taught us, rationality entails moral sense (Hume thought that humans have a natural desire to be social and benevolent). To refine the view using our concepts, perhaps we can say that evolution has endowed us with a module of narrow/minimal instrumental rationality. Our experience (and hopefully a good education from whatever society we happen to find ourselves in) then shapes domain-general mechanisms for what I like to call Kantian rationality: in the ideal, where our beliefs have to be maximally consistent, maximally philosophically grounded, and maximally sensitive to evidence.
I think most philosophers would be partial to this account of morality in humans (and if not, perhaps a sentimentalist view that makes the same appeal to domain-general mechanisms). Although, I’m not sure how many philosophers and psychologists will see it as fruitful to explain human morality with reference to other mammals. It has recently been found that bonobos will give up their meal to a bonobo stranger in exchange for social interaction. Frans de Waal has claimed that the foundations of human morality can be found in the animal kingdom: empathy and fairness. Capuchin monkeys demand fairness: they reject cucumber if the monkey next to them has been given a grape. But perhaps the right conclusion to draw from de Waals research is that mammals have a sort of sentimentalist proto-morality, whereas enculturated humans have a richer Kantian (rationalist) morality. This is certainly a conclusion that the Kantians John McDowell and Chris Korsgaard would be partial to, arguing that there exists a sharp cleavage between humans and other animals because humans have the unique, Kantian ability of being able to “step back” and assess the space of reasons. But, as Peter Godfrey-Smith has pointed out, humans are not the only animals that can assess the space of reasons. Greater primates can put off instant gratification for more substantial rewards later in time. So, perhaps the view that humans have only incremental improvements in the brain from their closest relatives is correct.
Moral nativism is the minority view, and hopefully for good reason. The one majority nativist view that is relatively uncontested is Chomskyan nativism: that language acquisition is a result of domain-specific module. The reason that it has been so uncontested, I think, is because it seems that language would give hominins overwhelming evolutionary advantage: it seems impossible that cooperation, long and medium term planning, and a myriad of other important behaviours could exist without language.
It would seem a wise middle-ground position in the nature-nurture debate to say that the mind is furnished with only domain-general mechanisms, except for a domain-specific mechanism of language acquisition (and perhaps instrumental rationality). But Prinz boldly argues that language may not be as native as has been supposed, pointing to recent innovations in our understanding of statistical learning and so forth. I think Prinz may be right in his assessment, but it certainly seems to me that there are native factors involved in our comprehension and use of language. Listen to a sentence, and then look at the sound wave it produces. The pauses we hear are no where to be found in the sound wave – a fact quite astonishing when you first see it. Our unconscious mind “cleverly” breaks up the language we are listening to into its understandable component parts. This is why it is near impossible to pick out the actual words in languages we are not proficient in. How is our mind capable of something so clever, if not for the fact that there are certain domain-specific mechanisms?
Once we are able to individuate words, how is it possible to learn their meaning? Perhaps by repeated ostensivity. But as Quine has taught us, reference is inscrutable: applying a word to an object is compatible with a myriad of different meanings. Say you visit a distant tribe, and you are trying to learn their language. You see the person from the tribe apply the word “gavagah” to a rabbit. However, you can’t say for certain that the tribesperson uses the word “gavagah” to mean rabbit. It may mean animal, or object in front of us, or even (fantastically) object x kilometres from the sun. So, it may be that something needs to be built in for us to know the reference of words, as it is underdetermined by what we can purely observe.
How do we define words? We define them by reference to other words, that are themselves defined. But what grounds this practice, if it is grounded at all? Bertrand Russell thought we at times need to rely on ostensive definition. But as Wittgenstein retorted, if we want to give an ostensive definition of the word “scarlet” for someone, they need to know we’re talking about a colour. Does this mean that certain rules of ostentation need to be built in?
Perhaps Prinz can give a hardcore empiricist account of all these considerations. What seems most mysterious to me is the evolution language itself, and its role in shaping the human mind (if it did). Philosophers of language, from Wittgenstein to Davidson and Dummett have been keen on the idea that language proceeds thought: thought is the internalisation of public language. Daniel Dennett has gone a step further and has argued that the human mind has been shaped by language. Language began amongst hominins by way of simple questions to one another. Thought began when one particular hominin asked a question and immediately received an answer – when no one was about.