On Empiricism
On Empiricism
What if empiricism had never been invented? It wasn’t invented till the seventeenth century: there is no trace of it in Plato and Aristotle, or their followers. It took a long time till philosophers got round to it; and it originated only in England not as a world-wide trend. It came from nowhere. Two individuals were responsible: John Locke and David Hume (Bishop Berkeley is a separate case). Locke believed in the tabula rasa conception of knowledge and held that all ideas derive from sense experience. Hume rejected the first doctrine but subscribed to the second. That doctrine proved to be of the essence. There were gaping holes in it from the start: the missing shade of blue, the problem of a priori knowledge, the inadequacy of mental images as an account of concepts, the lack of a coherent theory of abstraction. But I won’t go into these failings; my question is historical. Empiricism was a particular movement at a particular time and place; it wasn’t a universally accepted time-tested discovery. It went against the grain. It might never have existed. In a close possible world, empiricism never was invented—no Locke and Hume. Suppose this to be so: how would the course of philosophy look? It was a bad theory—what if we just delete it from history?
It took hold, we know that. It dominated English-speaking (and other) philosophy up until quite recently. Apart from John Stuart Mill and company, it shaped much of early twentieth century philosophy, overtly so. Russell was an empiricist. Logical positivism harked back to Hume (as Ayer explicitly stated). Husserl saw himself as a follower of Hume, as did Mach and Einstein. Knowledge was widely thought to be based on sense experience alone. True, there were dissenting voices, but they didn’t depart much from empiricist orthodoxy. Quine spoke of the dogmas of empiricism, but he still tied knowledge to the senses, now physically conceived (empiricism without experience). Husserl refined the empiricist conception of experience, but still clung to it. Ordinary language philosophy rejected the empiricist theory of meaning and concepts (ideas, images), but it still stuck to the observable phenomena of speech—what we hear rather than what we see (linguistic empiricism). It never abandoned empiricist epistemology in favor of rationalist epistemology. It celebrated common sense. The only examples of real resistance were Frege and Wittgenstein (early and late)—though even Wittgenstein followed the empiricist theory of necessity and the revolt against metaphysics. It was Frege who repudiated the whole empiricist worldview, notably in his attack on Mill’s theory of arithmetic, but also in his account of meaning. I can’t think of another major philosopher of this period who so clearly broke from the empiricist tradition. It was taken as a given. Strawson had empiricist leanings, as did Carnap, Ryle, Austin, Davidson, and others already mentioned; certainly, none of these advocated a return to rationalism. Chomsky did, but he is not a philosopher and never attacked the empiricist theory of a priori knowledge head-on. Despite its flaws, then, empiricism held sway throughout the twentieth century, though with minor modifications.
But I have not yet answered my question: what would philosophy have looked like if empiricism had never been invented? It is hard to say, because empiricism became so deeply entrenched. Frege never developed a general anti-empiricist philosophy, taking in ethics, philosophy of mind, and the nature of scientific knowledge (and non-scientific knowledge). I suspect ethics would have looked very different, because not in thrall to the empiricist theory of knowledge. Metaphysics would have been less stifled and apologetic. Philosophy of logic and mathematics might have been more central. But the main thing I think is that the mysterian viewpoint would have been far more salient, even orthodox. For the truth is that empiricism provided an impression of explanation of puzzling phenomena—a false impression, but an impression. It explains (allegedly) the origins of knowledge (concepts and whole propositions): it all comes from sensory experience. Nice and simple. Classical rationalism provides no real explanation—implanted by God is no explanation.[1] The empiricist explanation is also vaguely mechanistic: ideas causally derive from impressions; impressions cause ideas. This is supposed to be a law of nature, somewhat similar to Newton’s force laws: impressions have the power to produce ideas, as massive bodies have the power to produce motion (“ideational force”). What other explanation do we have? Empiricism gives us a natural science of knowledge formation: it is a matter of copying, imprinting. If this explanation is false, what do we put in its place? The mysterian answer is that we don’t know; so, it’s either a bad theory or no theory at all. Many people prefer a bad theory to no theory. What is clear is that in the historical absence of empiricism the mysterian position looms into view: knowledge is a mystery. I believe that much knowledge is innate, but I don’t think this is really an explanation of the origins of knowledge—for how does such innate knowledge come to exist in the first place? Empiricism explains origins (or purports to), but rationalism does not. In our alternative intellectual history, then, the focus is on resolving the mystery, or accepting it as insoluble. Empiricism purported to be the science of the mind analogous to Newton’s science of body, but it just isn’t a very good science. It is a kind of faith, whistling in the dark.
Let me put it even more bluntly: empiricism is a terrible theory cooked up by a couple of smooth-talking British blokes three hundred years ago, leading to the mess that was logical positivism.[2] That was repudiated in short order, but only minor amendments were made. We would have been better off without it. It still exercises a malign influence (particularly on scientists who are easily taken in by bad philosophy). But the alternative, supposing empiricism historically subtracted, is an absence of explanation, leading to a reluctant acceptance of mystery. Maybe the mystery would have been solved by now without the distraction of empiricism, or maybe it would not. In either case we would have been closer to the truth. A wildly speculative (and false) theory was converted into a dogma, and we have been living with the consequences ever since. The main dogma of empiricism is the doctrine that knowledge has anything essentially to do with experience.[3]
[1] I discuss this in Inborn Knowledge (2015). Exactly how is innate knowledge coded into the genes? It is—but how? And how did it evolve? This is close to the puzzle of the innate lexicon, emphasized by Chomsky.
[2] Moreover, Hume was badly misrepresented as more simple-minded than he was, as recent scholarship has demonstrated. A.J. Ayer’s interpretation of him was wide of the mark. We have been subjected to centuries of simplistic undergraduate Hume. Things are far more difficult than that.
[3] I haven’t tried to demonstrate this here, but I have written about it elsewhere (as have others). The essential point is that the deliverances of sense are never sufficient to generate knowledge, and are not always necessary for knowledge. Knowledge is a separate faculty from sensation; the former is not reducible to the latter. Also, sensation is a lot more complex than was traditionally recognized—more “cognitive”. Empiricism isn’t even true of the senses!
I agree that empiricism as the theory that all knowledge is based on experience derived from the senses is false. But you also write that “The main dogma of empiricism is the doctrine that knowledge has anything essentially to do with experience.” If we rejected the latter, I don’t see how natural sciences could exist.
It’s easy: information flow without consciousness. Think of a super-blindsight case in which information goes into the eyes and brain but there is no mediating conscious experience. Experience is just the contingent means of information flow. Remember subliminal perception experiments.
Locke and Hume’s writings sometimes demonstrate a sharp awareness of mystery. What was it about the problem of knowledge that led them to choose the simple, clear but untenable empiricist account over the less simple, mysterious, but ultimately much more plausible nativist explanation of the origin of knowledge?
That is an excellent question. They both were alert to mystery: causation in the case of Hume, the real essence of natural kinds in the case of Locke. But they never considered mystery as it applies to knowledge. Why, I don’t know. They seem quite inconsistent.
In the spirit of William James, there’s something of the Pragmatist to Empiricism- it is not just a philosophy of what is, but a way of learning about the world- in that it is an ideology of the scientific revolution- it is an article of faith saying that the world can be known through experience. Instead of asking what philosophy might have been for not Empiricism, let’s ask what science might have not been.
Tell me how wrong and misguided I am and why, and I’ll be a happy fellow.
Everybody thinks the world can be known by experience, even arch rationalists. The question is whether it must be known by experience.
Can we think thoughts that have absolutely nothing to do with the real world? That is a key question, that I have at least
Of course we can, though we can’t think thoughts with no content. Thoughts have to be about something, though it may be unreal, e.g. unicorns.
Unicorns might be considered having something to do with the real world, because they’re horses with horns, both of which are in the real world. Likewise a man who is 25 feet tall; we can think of a man in the real world and imagine him that tall. But can we think of anything that is not a variation of something in the real world but is totally unrelated to what we know of the real world?
This has no relevance to empiricism. What do you mean by the “real world’? Do you include numbers and logical relations, which are not objects of sense experience? If we are dreaming all the time, then none of our thoughts correspond to an external reality. Colors are often thought not to belong to the objective world, yet we think about them. It is not a necessary condition of thinking that the objects of thought exist (unless you go for some sort of externalism). These questions are not relevant to empiricism.
Freud famously goaded his reticent patients that surely you can’t be thinking about nothing so he’d agree for perhaps different reasons
Colors correspond in a contrived way to vibrations of light waves and correspond to the real world as much as langauge does to thought, am I not right, at least a little?
Color sensations are caused by wavelengths of light, but colors are not identical to them. They are secondary qualities. None of this has anything to do with empiricism.
Yes, you’re saying colors are a secondary property, that are not inherent to the physical world- I had that idea in grade school- but there must be a way around it- if language comes to reflect the properties it randomly correlates with why not color? Using Pavlovian association colors can be linked to other more real properties, with enough time and money and brainpower (which I might not have) a case can be made. Plus the qualitative variation of color can be perceived in the mind’s eye to correlate with the numerical variation of wavelength.
The subjective view of color occurs to almost everyone and is really a truism. Of course, everything mental is linked in various ways to the physical–so what?