The Return of Radical Empiricism


“All our knowledge begins with the senses, proceeds then to the understanding, and ends with reason. There is nothing higher than reason.” So wrote Immanuel Kant in his Critique of Pure Reason, one of the most influential philosophy books of all time. Kant is also the philosopher credited for finally overcoming the opposition between empiricism and rationalism in epistemology, as he realized that neither position, by itself, is sufficient to account for human knowledge.Kant was notoriously awoken from what he termed his “dogmatic slumber” [1] by reading David Hume, who had written in his Enquiry Concerning Human Understanding:“All the objects of human reason or enquiry may naturally be divided into two kinds, to wit, Relations of Ideas, and Matters of fact. Of the first kind are the sciences of Geometry, Algebra, and Arithmetic … [which are] discoverable by the mere operation of thought … Matters of fact, which are the second object of human reason, are not ascertained in the same manner; nor is our evidence of their truth, however great, of a like nature with the foregoing. … If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.”

The second part of the quote makes it clear that Hume, in turn, was reacting to the philosophical excesses of the Schoolmen, the medieval logicians who attempted to discover truths about the world by sheer power of mental analysis — an approach that, to be fair, goes back at the least to Plato himself, who was himself impressed by the effectiveness of mathematics in arriving at conclusions with certainty, and thought that the task of philosophy was to do likewise when it came to its own spheres of interest.

Why am I reminding you of all this? Because I am now convinced that we are witnessing a resurgence of what I call radical empiricism, the sort of thing that we thought we had left behind once Kant came onto the scene, and which, frankly, not even good ‘ol Hume would have endorsed.

Recently, here at Scientia Salon I published three essays — two by Robert Nola [2] and one by Coel Hellier [3] — that epitomize radical empiricism, more so in Hellier’s than in Nola’s case, I might add. Interestingly, Nola is a philosopher and Hellier a scientist, and indeed it is known by now that “scientism” — which is the attitude that results from radical empiricism — is being championed by a number of scientists (e.g., Lawrence Krauss [4], Neil deGrasse Tyson [5]) and philosophers (James Ladyman and Don Ross [6], Alex Rosenberg [7]).

Clearly, I find myself puzzled and bewildered by this state of affairs. As someone who has practiced science for a quarter century and then has gone back to graduate school to switch to philosophy full time I have a rather unusual background that, I think, makes me appreciate where radical empiricists come from, and yet which also precludes me from buying into their simplistic worldview.

In the remainder of this essay, then, I will try to do the following:

  1. Sketch out what I see are the logical moves attempted by radical empiricists;
  2. Show why they don’t work;
  3. Explain why this is more than an academic debate, and certainly more than “just semantics.”

Radical empiricists’ moves in logical space, and why they don’t work

My, by now, extensive readings of and conversations with radical empiricists have unearthed a number of standard moves they tend to make. I will briefly discuss six of them. Two obvious moves are (i) the use of an over-extensive definition of science and the assertion that other valuable disciplines — particularly (ii) logic and math — are “ultimately based” on empirical facts. Since radical empiricists do not seem to value (except for some degree of forced lip service when challenged) any other kind of inquiry or method of understanding (say, philosophy, literature, or the arts), it then follows that science really is all we should care about. It is as if they collapsed Hume’s already narrow distinction above between relations of ideas and matters of facts, arguing that the former are really a version of the latter anyway.

The concept of science, of course, has changed over time. The term did not actually exist as indicating a particular approach to knowledge of the world until recently [8]. Arguably, Aristotle (but not Plato!) was doing science, and so were some of the pre-Socratic philosophers, particularly the atomists. After the Renaissance, “natural philosophy” began to separate itself from philosophy more broadly construed, and finally a number of individual sciences became independent during the 18th, 19th and 20th centuries (most recently psychology, which was still a branch of philosophy until about the time of William James).

But modern defenders of radical empiricism don’t get to help themselves to the fact that what we understand by science has changed over the centuries, because if they did they might have to concede that, really, historically speaking it’s all philosophy.

Where could we turn for help, then? I’d say the dictionary, to get us started. Dictionaries are funny things. They play both a descriptive and a prescriptive role. They are descriptive of how — at any particular moment — a given culture uses a certain term; that, of course, can and does change over sufficiently long periods of time. But dictionaries are also prescriptive in the sense that, within a reasonably short time frame, they also tell us how we ought to deploy those terms. One doesn’t get to arbitrarily redefine words to suit one’s own ideological position or personal inclinations.

So, what are the dictionary definitions of science, mathematics and logic? Here they are (from my built-in Apple Dictionary):

science, the intellectual and practical activity encompassing the systematic study of the structure and behavior of the physical and natural world through observation and experiment. (Interestingly, the same dictionary also provides this alternative meaning: “knowledge of any kind,” but labels it as archaic.)

mathematics, the abstract study of number, quantity, and space.

logic, reasoning conducted or assessed according to strict principles of validity.

It ought to be clear even from these definitions — which are congruent with the vast majority of the specialized literature on the philosophy of science, of math, and of logic — that mathematics is distinct from but akin to logic, and that both of them are very distinct from (although very useful to) science. Hume was onto something, after all.

As I mentioned, the most common refrain from radical empiricists when faced with the above is that math and logic “ultimately” are rooted in empirical knowledge, a recurring example being that we believe that 1+1=2 because we can see that if we put side by side two objects of the same kind we get a total of two objects of the same kind. Another example is that standard practices in logic, say modus ponens [9] are adopted because they “work” in the real world.

Both responses miss the mark because they subtly but surely change the conversation. The first example tells us at most that human beings began to think about abstract objects prompted by elementary empirical observations. But the question at hand is not how mathematical reasoning originated in the Pleistocene, it is what kind of mental activity is modern mathematics. And much of it has nothing whatsoever to do with empirical groundings of any sort. Yes, math is deployed as a tool in science and in all sorts of other applications, but there are huge swaths of mathematical territory that neither describe anything in the world nor are pursued by mathematicians for any practical reason at all.

As far as logic is concerned, a similar reasoning holds there too. And the example of the utility of modus tollens is another red herring that derails the conversation: the question isn’t whether some principles or methods of logic are useful and therefore employed in other areas of application. Of course they are. But logicians — just like mathematicians — are concerned with the formal structure and internal coherence of their constructs, not with whether they do or do not map onto the real world. Many of those structures do not, in fact, map onto the world. When they do, it is only because the world as it actually is does not contain logical contradictions and mathematical inconsistencies, so math and logic are bound to describe the real world together with countless other hypothetical ones (this is true quite irrespective of the ontological question concerning abstract objects, i.e., regardless of whether one is inclined to be a Platonist or not).

Another common move employed by radical empiricists is to (iii) deny the existence of a priori knowledge. It cannot exist, because otherwise they’d have to admit that science (understood as an essentially empirical enterprise) isn’t the source of all knowledge. The most sophisticated of the new wave of radical empiricists sooner or later will cite W.V.O. Quine’s famous rejection of the difference between analytic (a priori, by reasoning) / synthetic (a posteriori, by observation) truths in his paper, “Two dogmas of empiricism” [10]. But I bet that a good number of them have not actually read it, and even more likely that they are not aware of the criticism it got and of the significant amount of backtracking Quine himself had to do throughout the rest of his career.

You see, Quine made ample room for a priori truths in his “rejection” by acknowledging two things: the special status of mathematics as a type of science because it has applications in science (but see above for why this is irrelevant), and the fact that tautological statements (the famous “bachelors are unmarried men” kind of thing) are indeed examples of analytic truths, but turn out to be “epistemically insignificant” according to Quine’s judgment [11]. Well, that’s his opinion, and given that much of logic and math are built on tautologies, a very debatable opinion at that.

A better example of what Quine was talking about are equations such as F = ma from Newtonian mechanics. He thought that this may look like an analytic truth, specifically a definition (hence tautological) of force. But in fact the equation is only true within a specific empirically-based theory of the natural world, its truth not deriving from mathematical reasoning per se. I have no qualms with that, but acknowledging this is a far cry from saying that there are no a priori truths and no difference between synthetic and analytic statements.

Radical empiricists’ next move is to (iv) point out that science uses the same fundamental tools — observation and reason — that we all deploy in everyday life whenever we want to know anything at all. This is just as true as it is utterly uninteresting. It would be surprising, in fact, if science as a human epistemic activity were to somehow transcend the basic intellectual faculties of our species and operatesui generis (just as it would be equally surprising if there were a philosophical method that was entirely distinct from normal human reasoning). Of course doing math, logic, philosophy, art, literature, navigating the New York City subway system, and plumbing use facts (whenever appropriate) which are analyzed by reason. Nevertheless there are tons of interesting distinctions among all those activities, distinctions that are lost by the quest for what I have come to call “explanatory monism,” the obsession with a one-size-fits-all epistemology. Epistemic pluralism is much more interesting and fecund, not to mention more accurately reflective of actual human practice.

The next move, then, is a partial retreat on the previous one, and goes something like this: (v) there are no sharp distinctions between the mentioned activities, so there is no principled way to distinguish among them. To which I can only reply in two ways: there is no sharp distinction separating a helicopter, a jumbo jet and a Saturn rocket, as they are all flying machines. But if you think there are no interesting differences among them you are sorely mistaken. Also, anyone seriously arguing that philosophy, math, logic and, say, biology, are more or less the same thing has clearly not read a single technical paper in more than one of those disciplines.

There is one more defense of radical empiricism, rooted in a kind of greedy reductionism: (vi) the idea that “ultimately” whatever it is we are interested in (poetry, art, mathematics) is made of physical matter or done by beings made of physical matter, so that it all comes down to neuroscience or, if the radical empiricist is particularly bold, to quantum mechanics.

This, again, is a move predicated on shifting the discourse without apparently realizing that one has done so. The issue isn’t what something is made of (ontology), but rather how we may best proceed in understanding it (epistemology). Epistemologists understand very well that for any particular problem X there is a usually small number of levels of analysis that are most informative and appropriate in order to understand X. These can be located one or two (loosely defined) levels of complexity below or above X itself, but the explanatory returns taper off very quickly after that. Let me give you an example.

Let’s say you want to understand the population dynamics of a species of plants, for instance belonging to an invasive species (this comes straight out of my work as an empirical scientist, as you might have guessed). It is of no use to point out that plants, “ultimately” are made of quarks. A quantum mechanical theory of population dynamics — even if possible in principle — is never going to be developed and it wouldn’t help anyway because it would be far too complicated (and unnecessarily so) for a human to comprehend. Instead, the population biologist looks at population genetics (circa one level of complexity below that of organismal biology) and at ecosystem theory (circa one level of complexity above).

Similarly, it is a good bet that to understand economies one needs to operate at the level of economics as autonomous science, plus at the levels of, say, human sociology and psychology. Neuroscience is not likely to be helpful, because it would be too detailed for the problem at hand, even though of courseeconomies are inventions of the human mind, and of course the human mind is the result of the activity of the brain, and of course the brain is made of neurons and other cell types. If you are not convinced, try to go even further down the hierarchy of complexity. How likely is it that we could develop a useful theory of economies based on molecular biology (after all, the brain is made of molecules!)? What about fundamental chemistry (those molecules are made of atoms!)? And so forth until we get to the single wave function that allegedly represents the entire universe.

So, a crucial reason to maintain distinctions among fields of inquiry — even when acknowledging bridges, cross-pollination, and similarities — is that ultimate reductionism will always be a losing epistemic proposition, even if one agrees with the ontological statement that everything is made of quarks (or strings, or wave functions).

Why bother?

I find all of the above intrinsically interesting as an example of intellectual debate about matters of proper definitions, conceptual understanding of different human epistemic activities and so forth. In other words, as a professional philosopher this kind of discussion represents a worthwhile venture into the philosophy of science and in epistemology. But there are far more practical reasons why the assault of the radical empiricists ought to be resisted.

Two reasons in particular are of concern to me: the damage being done to non-scientific disciplines, and the damage potentially to be suffered by science itself.

For years now the humanities and any non-STEM (Science, Technology, Engineering and Mathematics) fields have been in retreat in colleges throughout the world, especially in the US. This retreat is the result of a number of factors, perhaps foremost among them the increasing importation of business-style models into academia and the resulting conviction that if studying a given discipline doesn’t have an immediate payoff in terms of employment then it is not worth studying. This is a false and perniciously instrumental view of higher (and lower, really) education, which has the potential to undermine people’s ability to develop into cultured human beings capable of reflecting on what they do, how they do it , of appreciating all aspects of life (not just jobs and livelihood), and of making informed decisions as members of a democratic polity.

The aggressiveness of radical empiricists and their dismissal of non-scientific fields exacerbates this problem, and in my mind, therefore contributes to undermining the very fabric of our democracy and to decreasing the quality of our life.

This may sound like “defending the turf,” and in a sense it is. But some turfs are worth defending against an all-encompassing cultural imperialism that risks to flatten the intellectual landscape in the name of Science (notice the capital S). And no, I’m not at all coming at this from the point of view of mystical or theological woo in constant entrenchment against science — as I hope is abundantly clear by the body of my writings.

The second worry may seem specious, but I think it is just as important to appreciate. I think that an over-emphasis on the powers and overall reach of Science will, in the long run, do harm to actual, good science. We are already facing a public that is increasingly unwilling to trust scientific findings (just think of the widespread rejection of the theory of evolution or the notion of climate change, or of the uncritical acceptance of a non existent causal link between autism and vaccines, to mention just a few examples). The more scientists are seen as arrogantly dismissive of any other dimension of human experience the more this distrust will grow and fester. And science, the real science done in countless laboratories and university centers across the globe, is just too precious an achievement of humanity to let it be damaged by an emotional reaction to the loud, radical statements of an overbearing but comparatively small number of highly visible public figures.

Isaac Asimov famously said that “The saddest aspect of life right now is that science gathers knowledge faster than society gathers wisdom.” Indeed, but we don’t get wisdom from science alone.


Massimo Pigliucci is a biologist and philosopher at the City University of New York. His main interests are in the philosophy of science and pseudoscience. He is the editor-in-chief of Scientia Salon, and his latest book (co-edited with Maarten Boudry) is Philosophy of Pseudoscience: Reconsidering the Demarcation Problem (Chicago Press).

This post originally appeared here:

[1] As he put it, in Prolegomena to Any Future Metaphysics.

[2] Scientism: ‘Yippee’ or ‘Boo-sucks’? — Part I and Part II, by Robert Nola, Scientia Salon, 18 and 19 August 2014.

[3] Defending scientism: mathematics is a part of science, by Coel Hellier, Scientia Salon, 21 August 2014.

[4] Lawrence Krauss: another physicist with an anti-philosophy complex, by Massimo Pigliucci, Rationally Speaking, 25 April 2012.

[5] Neil deGrasse Tyson and the value of philosophy, by Massimo Pigliucci, Scientia Salon, 12 May 2014.

[6] James Ladyman on Metaphysics, Rationally Speaking podcast, 9 September 2012.

[7] Is science all you need?, by Massimo Pigliucci, The Philosopher’s Magazine, 2nd Quarter of 2012.

[8] The very word “scientist” was coined by philosopher William Whewell in 1833, in response to a challenge issued by poet S.T. Coleridge.

[9] Modus ponens.

[10] Two dogmas of empiricism, by W.V.O. Quine.

[11] Quine, W.V.O. (1991) Two Dogmas in Retrospect. Canadian Journal of Philosophy 21:265-274, see p. 271.