Monday, September 5, 2016

Innatism and the justification of philosophical method

The traditional form of a quintessentially philosophical question is: What is X?  As in: What is justice? What is knowledge? What is life?

These questions are traditionally associated with a particular form of inquiry.  Roughly:
  1. Think very hard about different states of the world in which X is clearly present. 
  2. Detect some property (or set of properties) P shared by all such states. 
  3. Propose P as necessary and sufficient for X. 
  4. Submit your proposal to other philosophers for rigorous cross-examination. 
Theoretically, this process, iterated, eventually reveals the true nature of X.

I say 'theoretically' because we are entitled to ask why. What theory of inquiry motivates the view that we can learn substantive truths about the world just by reflecting on the content of own minds?

One possible reply is this: We don't need no stinking theory of inquiry. We just need to point to a heap of solid results produced thereby. Right. Well, there's the rub.

Plato, who raised this form of inquiry to high art, seemed to recognize that this question requires an answer. In struggling with it, he was led to one that is as beautiful as it is absurd: Our souls contain every truth worth knowing, but we forget them during the trauma of birth. On Plato's view, what we call learning about the world is really a process of recollection. This is his theory of anamnesis, which leads Plato to endorse an ancient myth of reincarnation.

Plato's basic answer persisted in one form or another as the doctrine of innate knowledge. Two thousand years later we find a fellow traveler, René Descartes, smack in the middle of the scientific revolution, self-consciously defending the same basic method in his Meditations on First Philosophy. Descartes famously argued that the Creator outfitted our intellect with innate indubitable knowledge of the kind of world we live in as well as reliable resources for learning all about it.

Usually we characterize the rationalist commitment to innatism by reference to the view that genuine knowledge is incorrigible, i.e., true beyond all possible doubt. Rationalists like Plato and Descartes believed that only the exercise of pure reason could produce such truths. But empiricists like Aristotle and Hume accepted the incorrigibility of genuine knowledge as well. What they hated was the conclusion that to know we know anything about this world we must presuppose a completely different one.

So a commitment to innatism has two main sources. One is the view that empiricism is too impoverished a framework to explain the possibility of human knowledge. The other is that it justifies the traditional method of philosophical inquiry.

Maybe now I will say this? Our experience with science has taught us that empiricism is right and innatism is wrong. So it's high time we exchanged armchair philosophical methods for more enlightened naturalistic ones. Well, there are plenty of philosophers today who do say this. I tilt that way. But it is too easy. Let's see why.

First, let's be clear that empiricism has, for the most part, won. But that is not because classical empiricists succeeded in showing that the ultimate foundation of infallible knowledge is experience. It is because science, using a fundamentally empirical method, has amassed a magnificent mountain of knowledge in the absence of any such assurances. By brute force, the success of science has ushered in an age of fallibilism, which is the idea that we can come to know X through methods that do not guarantee the truth of X.

Epistemologically speaking, this is just a whole different world. Back in the day, empiricists and rationalists disagreed on the ultimate foundation of knowledge, but they totally agreed that, whatever its source, the method for producing knowledge had to assure certainty. It just couldn't be any other way. But they were all wrong. Scientific knowledge is not deduced from first principles. It originates in guesses, hypotheses that attempt to account for why we observe the world behaving as it does. Hypotheses that survive extended, merciless testing get promoted to theories and may ultimately earn the status of knowledge. But they never get tenure. All scientific theories remain subject to performance review and none rise completely above their uncertain provenance.

Second, innatism persists within science itself. Rationalists were actually correct that the mind can not just be a "blank slate" at birth. In order to be capable of gaining knowledge about the world, it must begin with some kind of basic structure. Scientists differ with respect to the content and plasticity of these presuppositions, but nobody represents them as incorrigibly correct. Scientific innatism explains how infant brains can develop into functional adult ones, not how infant brains come into this world containing fundamental truths of the universe.

Third, at various stages of inquiry scientists, too, pose questions of the form "What is X?" and, like philosophers, they must consult their own minds for the answer. But they approach these questions a bit differently than we do.

First, the intuitions they consult are not ordinary intuitions, but those of a specialized research community trained up on a technical vocabulary and specific methods of inquiry.

Second, scientists do not believe that thinking long and hard on a question like "What is a hydrogen bond?" is sufficient to uncover its true nature. They believe that only experiment can reveal that.

Third, this kind of thinking does not occur in a vacuum, but with specific explanatory goals. When Erwin Schrödinger posed the question "What is Life?" he was not taking a holiday from physics to do a little philosophizing. Rather, he was trying to bring living systems within the explanatory purview of his own discipline.

Finally, scientists don't normally try to provide a set of necessary and sufficient conditions for X-hood, and they typically don't get their panties in a bunch over a single striking counterexample to an otherwise useful definition. Scientists work in a manifestly messy world and they have learned how to produce in the face of all manner of uncertainty.

G. Randolph Mayes
Department of Philosophy
Sacramento State


  1. Hello Professor Mayes,

    Thank you for writing this article :).

    Do you think that a commitment to innatism requires commitment to us having privileged access to contents of our mind? If not, what would innatism look like without privileged access?

    Do you think we have privileged access to the contents of our mind, such as our beliefs? If not, why do you think so many of us imagine we do and what do you think is the relationship we have to these contents which accounts for why we do not have this access?

    Hopefully my questions are not littered with misused technical terminology, but if anything is unclear let me know and I'll elaborate.

    Thank you very much,
    Stan Lovelace

  2. Hi Stan, nice questions.

    I think the answer to your first question is no, but that they do go together well. I think privileged access is a Cartesian notion that you don't really find much prior to the enlightenment. This is partly why you see Plato representing philosophy, not as a meditative exercise, but as an inherently group endeavor. Socrates is always pestering people to think with him. In Meno, the slave boy needs Socrates' prompting to unlock what is already inside of him. Privileged reflective access to our innate ideas is the specifically Cartesian solution to how our knowledge can be infallible.

    I think we have privileged access to our own mental states in the sense that each one of us has a unique first-person perspective on them. The idea of privileged access also tends to suggest that it is a more reliable channel of information than external observation of our behavior or (in the case of neuroscientists) brain states. This view is easy to motivate with well-chosen examples. It seems that I must know better than you do what number I am thinking of right now, for instance. But it's not at all clear that I know better than you do what kind of food I prefer or whether I am afraid of ghosts. Right now we really don't understand how introspection works. As far as why we imagine that we have privileged access (in the reliability sense) I don't know. I think it is partly that we are children of Descartes, partly that it is pride of ownership, and partly that we are generalizing from a non random sample that tends to be associated with a lot of subjective certainty. In this sense I think it's a lot like the question why people are so confident in the reliability of eye-witness testimony even though the data suggest that it is one of the least reliable kinds of evidence.

  3. Hi Randy. Just a quick (and perhaps not very interesting) comment.

    It seems to me that in Descartes, innateness and indubitability are not coextensive. I know mathematical truths innately, on his view- e.g. the Pythagorean Theorem. However, I can doubt these things under the Evil Demon hypothesis.

    This gets complicated quickly, with discussion of the Cartesian Circle; just what, exactly, does Descartes intend to doubt under the hypothesis of the Evil Demon? But at the very least, he doesn't seem to want to say that all innate propositions are indubitable.

    I don't think this affects your larger point so it may seem quibbling... my apologies. But discussion is good.

  4. David, thanks, yes I think you are completely right about that. I'm obviously not a Descartes scholar, but it seems to me that there are many things that turn out to be known with certainty, even though they can initially be doubted. This would include God's existence, too. But he then establishes God's existence with certainty and then it seems to me the certainty of all of our innate ideas, since He put them there and He is good. But you're right that this is a derived conclusion for Descartes, it does not follow from the fact that they are innate that they are indubitable.

  5. Here are my two cents (from one who doesn’t work in epistemology).

    I think I’m in agreement with you that empirical methods are wonderful, but we shouldn’t be too hasty to rule out innatism.

    Methodological/procedural assumptions also require justification, but they generally aren’t placed under the same scrutiny as other substantive assumptions. Our empirical conclusions are often the product of our initial assumptions and the design of a study, which relies on one or more methodological assumptions. Methodological or procedural assumptions, however, generally aren’t placed under the same scrutiny as other substantive assumptions. Many of our basic methodological assumptions, just like many basic substantive, especially metaphysical, assumptions, cannot be proven true or false demonstrably. I would be careful to rule something out from the outset without such proof.

    I’m sympathetic to the rationalist who accepts knowledge based on experience and pure reason. The rationalist, as I understand her, is not rejecting empirical methods, but is saying that knowledge can be obtained through both experience and pure reason. The rationalist also may say that one is a higher form of knowledge (based on some standard of knowledge), but we can set this further claim aside.

    I think there’s good reason to be open to knowledge by pure reason alone. You mentioned one. We may be incapable of gaining knowledge without some basic structure. Our knowledge isn’t comprised of raw data, but the data is organized in complex and intricate ways through the use of categories (distinctions) and concepts. Some of this may be the result of the brain and its processes, but some of this remains to be explained. Animal brains, including ours, view the world very differently--why we view the world as we do, and possibly in ways more advanced than other animals, is a bit mysterious. It seems our brains come ready-equipped with capacities, categories, and concepts (not to mention a backdrop of consciousness). Again, we may be able to tell some evolutionary story to explain some of this, but maybe not all of this.

    The other reason that comes to mind is that the scientific method can prove whether hypotheses are true or false, but it does not explain the origin of our hypotheses or “guesses.” Proven hypotheses comprise our knowledge, but their origin also remains a bit mysterious. Sometimes the guesses may be pure guesses. Or, and this is probably more often the case, they may be based on keen observations of what we already know. But then even our keen observations of what we already know may be divided into two parts: (1) what we already know and (2) extensions based on what we already know. It seems to me that (2) involves some additional thought or intuition. One explanation for this is that our hypotheses, or at least some of them, include some pure thought or intuition. I’m not suggesting that the alternative explanation can be proven true, but I am saying that when there are alternative methodological assumptions, neither of which can be proven demonstrably true, this at least gives us good reason to exercise intellectual humility.

    We’ve certainly amassed a great deal of knowledge through science, but I don’t know if we know all that we know through science (or empirical methods) alone. The empiricist seems to be making the bolder claim that knowledge is obtained through experience alone, rejecting knowledge from pure reason. This may be the popular view today, but it involves a methodological assumption that cannot be proven to be true demonstrably (if a person accepts both, the fact that one is fertile, doesn’t show that the other is impotent).

    1. Chong, thanks for all of this. I definitely agree with you that today the categories of "empiricist" and "rationalist" have changed dramatically, and that science as practiced today can be reasonably construed as a kind of synthesis. Today we call a scientist like Chomsky a rationalist mostly because of how much innate structure he presupposes in explaining the possibility of language acquisition, but not because he eschews empirical methods.

      Similarly, today's empiricist is not someone who believes we acquire knowledge from experience alone, but rather someone who thinks we can presuppose a great deal less innate structure to explain how we learn about the world. (Just as Darwin was able to presuppose a great deal less structure to explain the origin of life than, say, Paley.)

      Some empiricists today, Gopnik, for example, believe that our DNA encodes certain kinds of basic theories about the world, or at least an innate tendency to develop them. But what I think hardly anyone argues is that they are therefore indicative of the way the world is. At best, they are just a starter pack of heuristics and folk theories that have a certain amount of survival value for helping infants to control their parents and participate in the society that makes their continued survival possible.

      Some of what you say, I don't follow completely. For example, I don't know what you mean when you say that some evolutionary story may explain some of our cognitive capacities, but maybe not all. What other kind of story could there be?

  6. Randy,

    Judging by just the replies so far, you've given us all a great opening dance sequence this fall. But hey, that may just be my empiricism talking here based on what I've seen up until this point.

    More seriously, I'm going to wonder out loud here whether the historical representatives of innatism that you cite might actually be actually just the sort of necessary intellectual partners with the empiricists you see as refuting them through honest toil rather than theft…and not just intellectual partners as a foil, but as those rowing in the same direction.

    How so? Well, take the simplest sort of cartoonish experiment: I will do X to see if Y or not-Y is the case.

    Where do we get the idea that X can decide whether Y or not-Y, if not…wait for it…from somewhere innate?

    I suspect this question has a long history of being asked, and answered, in more or less sophisticated ways than I just asked it.

    But I guess it's like a twist on the turtle question: is the good science, the contemporary science, the empirical science, really built on non-innate ideas all the way down?

  7. Russell, thanks for this. Funny, an earlier incarnation of this post was titled "It's theories all the way down". The argument was going to be, essentially, that there is no such thing as a truth that is simply handed to us, that all of our beliefs are subject to refutation and revision, including those ideas that seem to be just part of our cognitive apparatus or necessarily so. But then our thousand word limit kicked in with a vengeance so I think that will be a future post beginning with a reminder of this one.

    I think the beginning of an answer to your question is this: Our brains have some innate structure, but it is a great deal less than previous thinkers had thought necessary. This is a recurring theme in science, most obviously (as I noted in my reply to Chong) in Darwin's theory of natural selection where it has been shown that incredibly sophisticated design can emerge, not from a designer, but from a few simple principles: random mutation, transmission, selection. We also see it in explaining other emergent behaviors, such as that incredible dance or murmuration of starlings each one of which is just following a few simple rules for flying near their neighbors.

    The rest of the answer awaits the results of a lot of empirical research. There is plenty to suggest that children have an innate primitive grasp of logic, causality, probability and even morality. None of this is problematic from an empiricist perspective as long as we are committed to discovering a good non magical (i.e. phylogentic and otogenetic) theory of how this happens.

    The main point is that there is an enormous gap between acknowledging that some of our ideas are innate and concluding that they they are a guide to the structure of the universe.