Against Empirical-ish Philosophy: Reply to Mizrahi, Jeroen de Ridder

Is the debate about scientism ‘a battle for the soul of philosophy’, as Moti Mizrahi (2019) contends? If it is, it is unlikely that the proponents of scientism would be inclined to put it this way … [please read below the rest of the article].

Image credit: Daniel Friedman via Flickr / Creative Commons

Article Citation:

de Ridder, Jeroen. 2019. “Against Empirical-ish Philosophy: Reply to Mizrahi.” Social Epistemology Review and Reply Collective 8 (12): 8-12.

🔹 The PDF of the article gives specific page numbers.

Articles in this dialogue:

Is the debate about scientism ‘a battle for the soul of philosophy’, as Moti Mizrahi (2019) contends? If it is, it is unlikely that the proponents of scientism would be inclined to put it this way. Being the empirically undetectable immaterial entities they are, souls typically don’t make the cut in properly scientistic ontologies. A more scientism-friendly paraphrase of Mizrahi’s question might be to ask whether the debate about scientism is a battle for philosophy’s DNA, or perhaps the memetic equivalent thereof.

The Incompatibility of Scientism and Philosophy

I do think Mizrahi is on to something. Suppose we understand scientism along the lines of Mizrahi’s Weak Scientism, i.e., the claim that “of all the knowledge we have, scientific knowledge is the best knowledge” (Mizrahi 2017, 354), or a stronger claim to the effect that “science is the only source of justified belief or knowledge about ourselves and the world” (De Ridder 2014, 25, cf. also Peels 2018). As I’ve argued a few years ago in this journal (De Ridder 2014, 27–29), this understanding of scientism straightforwardly implies that we ought to dismiss philosophical theorizing on a topic as soon as science begins to investigate it, because much philosophical theorizing makes ineliminable use of armchair methods—conceptual analysis, a priori reasoning, thought experiments, intuitions, normative reflection, common sense, or combinations thereof—rather than empirical scientific methods. Hence, in so far as one takes these or similar a priori methods to belong to the essence or soul of philosophy Mizrahi is right that scientism poses a threat to the legitimacy of philosophy, at least in its traditional shape.

One might think that this pretty much settles the question that is the title of Mozrahi’s essay—whether the scientism debate is a battle for the soul of philosophy. No doubt, there is room for fine-tuning and chisholming by considering different versions of scientism, different construals of a priori philosophical methods, and different broad conceptions of philosophy in order to explore their mutual compatibility and interrelations, but the basic insight seems solid. Preferential or exclusive reliance on empirical scientific methods leaves no room for a priori philosophical inquiry. So what work could there be left for empirical methods in addressing Mizrahi’s question? If we’re really strict: none. The quick a priori argument from the previous paragraph settles it.

Empirical Questions About Philosophy

There are of course plenty of neighboring questions which do lend themselves to empirical inquiry. To mention just a few: How many philosophers subscribe to the self-image of philosophy as (at least in part) essentially involving a priori inquiry? How do philosophers feel about the potential uses and benefits of employing empirical methods in philosophy, alongside a priori methods or instead of them? What are best (and worst) practices for using empirical methods in philosophy? What portion of philosophy publications rely exclusively on a priori methods? Is the use of empirical methods in philosophy on the rise and can one discover patterns or trends over time? And so on. Indeed, Mizrahi’s own two hypotheses are also relevant. Paraphrasing: do philosophers feel threatened by scientism because they see it as a threat to (1) the future of academic teaching of philosophy and/or (2) the soul or essence of philosophy as an a priori discipline?

These and similar question clearly require empirical methods: quantitative surveys, qualitative interviews, bibliometric analyses, perhaps automated content analyses of papers and abstracts, and sound statistical analyses. It’s not immediately obvious that such inquiries belong to the core business of the academic discipline of philosophy. Perhaps a more natural home for them would be psychology, sociology, or anthropology, or, for large-scale bibliometric analyses and automated content analyses, digital humanities or meta-research.[1] But, since none of these fields currently devote much attention to philosophy, maybe philosophers themselves should take the lead in putting empirical questions about their discipline on the agenda, at least for the time being.[2] Ultimately, however, I don’t think discussions about whether such projects ‘really’ belong to philosophy matter much. When practitioners of an academic discipline spend lots of their time on boundary-keeping that usually isn’t a sign of strength. Better to focus on formulating and answering worthwhile first-order questions.

Questions for Mizrahi

So, while I think Mizrahi is doing us a service in putting questions about the psychology and sociology of philosophy on the agenda and exploring the use of bibliometric analysis, the execution of his project leaves a lot to be desired. If we want to get serious about employing empirical methods to answer questions about the practice of philosophy, we need to shape up methodologically. Let’s consider some problems.

1. Formulation of Hypotheses

Both hypotheses speak of ‘many philosophers’. This doesn’t lend itself to empirical testing without further specification of how many will count as ‘many’. Is ‘many’ supposed to be an absolute number (more than 250 philosophers with academic employment?) or a relative one (more than 10% of philosophers?).

2. Mismatch Between Hypotheses and Methods

Both hypotheses ostensibly concern the beliefs or feelings of philosophers: philosophers ‘find scientism threatening because they see it as a threat to …’. To learn about other people’s beliefs and feelings, or mental states more generally, we need to ask them. Survey methods and interviews would thus have been the most appropriate methods here. When this is impossible or too burdensome, we have to make do with observing people’s behavior and inferring conclusions about their mental states from that, but this often introduces considerable uncertainty, as behavior can be caused by many different combinations of conscious and unconscious mental states.[3] Bibliometric analyses of the sort Mizrahi has carried out are an example of observing behavior. But they are a very blunt tool. Publishing papers is an extremely complex sort of behavior and doesn’t have a straightforward connection to any unique combination of beliefs, desires, or feelings.

3. Failure to Rule Out Alternative Hypotheses

This brings me to a third worry. Mizrahi doesn’t do anything to control for confounding variables, or, in other words, to rule out alternative explanations that might account at least equally well for the observed rise in publications mentioning scientism. He takes the number of publications mentioning scientism as a direct and unequivocal indicator of philosophers and religious scholars feeling under threat, because “when scholars are concerned about something, they write about it” (Mizrahi 2019: 3).

While I don’t think it’s false that academics sometimes write about things they feel threatened by, there are plenty of other possible reasons why they may want to write about scientism. Without going out on a limb here, I suspect they might sometimes think the topic is genuinely interesting and worthy of scholarly attention. Alternatively, perhaps they need a few publications to get tenure and they see low-hanging fruit that is ripe for the taking. Maybe they read a scientistic popular science book (of which there seems to be no shortage) and felt the need to address the implicit epistemology therein. Nothing in Mizrahi’s data suggests that any of these other hypotheses are false.

If I had to make a guess, I think the explanation for the rise in publications mentioning scientism is actually even more prosaic than the suggestions above. It’s a well-confirmed trend in academic publishing that we do more and more of it. Estimates over the past decades vary between a 5–6% (STM 2018) to an 8–9% (Landhuis 2016) yearly increase in the number of scientific papers that are published. ‘Publish or perish’ culture may have been most pronounced in the biomedical and social sciences, but humanities disciplines, too, have seen considerable growth in the number of papers and books being published.

When we correct for this general trend, it’s unclear that Mizrahi’s numbers show anything significant. In absolute numbers, we could probably find an increase in publications on any topic of some relatively permanent interest to philosophers. To support his hypotheses, then, Mizrahi should have shown that there is a relative increase of publications about scientism: if the growth in publications mentioning scientism outstrips the growth of philosophy publications generally, that would be an indication that scientism has become a topic of some special concern to philosophers (although we would still need more evidence to determine what’s behind this increased attention).

4. Correlation Doesn’t Imply Causation

Mizrahi’s use of regression analyses is also premature. Regression analyses are useful when we have good reason to believe that causal relations obtain between variables. Mere correlations are not a good reason to assume causality; they are at best defeasible evidence for causality. Purely observational (rather than experimental) data about numbers of publications of the sort Mizrahi has collected, does nothing to make a causal relation plausible. So when Mizrahi describes the results of his regression analysis in causal language (e.g., “this suggests that the number of philosophy publications that contain the phrase ‘experimental philosophy’ explains 72% of the variation in the number of publications that contain the term ‘scientism’” (ibid.: 7, my italics)) this is simply unwarranted. Nothing in his data suggests a causal relation between the two variables. And indeed, as we saw above, there are a number of other plausible explanations on offer.

To drive the point home, consider an example. Alvin Plantinga published his modal ontological argument for God’s existence in 1974 and Hilary Putnam first proposed the no miracle argument for scientific realism in 1975. Bibliometric analysis of philosophy publications from 1975 onwards will most likely show at least a decade of increasing numbers of publications mentioning ‘ontological argument’ and ‘no miracle argument’. Running a regression analysis on this data might well suggest that a very high percentage of the variation in numbers of publications mentioning ‘no miracle argument’ is ‘explained’ by the variation in the number of publications mentioning ‘ontological argument’. Have we now unearthed a previously unknown causal connection between the philosophy of religion and the philosophy of science? Of course not. We’ve merely latched onto a spurious correlation. I suspect that this might really be what’s going on in Mizrahi’s analysis.

My earlier point about the increasing numbers in scholarly publications generally further supports this suspicion. If we take any two philosophical topics that have either attracted increasing attention or at least have witnessed a relatively stable interest among philosophers, we should expect to find increasing absolute numbers of publications over the past decades. Hence, we could find relatively strong correlations between the numbers of publications about these topics and a regression analysis would seem to suggest that a substantial part of the variation in the one number is explained by that in the other.


Mizrahi’s recent papers have done a lot to advance our understanding of scientism. And his most recent contribution is to be commended for putting empirical questions about philosophers’ underlying feelings and views about scientism on the agenda. As I’ve said above, there are many worthwhile questions in the psychology, sociology, and anthropology of philosophy that can be fruitfully addressed by the sophisticated methods employed in these disciplines. I fully agree with Mizrahi that the use of empirical methods in philosophy (or, perhaps more accurately, in the empirical study of philosophy) can be useful.

However, if we as philosophers want to take this task seriously, we ought to study empirical methodology carefully. We should definitely not try to wing it. I’m afraid Mizrahi may have underestimated things a bit in this regard. As I detailed above, his methodology suffers from a number of debilitating problems. As a result, neither the analyses nor the conclusions he draws from them have much going for them. Clearly, this is something we should avoid. The empirical study of philosophy deserves better than mere ‘empirical-ish’ philosophy.[4]

Contact details: Jeroen de Ridder, Vrije Universiteit Amsterdam,


Bourget, David and David Chalmers. 2014. “What Do Philosophers Believe?” Philosophical Studies 170 (3): 465–500.

De Cruz, Helen. 2017. “Religious Disagreement: A Study Among Academic Philosophers.” Episteme 14 (1): 71–87.

De Cruz, Helen. 2018a. “Prestige Bias: An Obstacle to a Just Academic Philosophy.” Ergo 5 (1): 259–287.

De Cruz, Helen.  2018b. “Religious Beliefs and Philosophical Views: A Qualitative Study.” Res Philosophica 95 (3): 477–504.

De Ridder, Jeroen 2014. “Science and Scientism in Popular Science Writing.” Social Epistemology Review and Reply Collective 3 (12): 23–39.

Graham, George. 2019. “Behaviorism.” In The Stanford Encyclopedia of Philosophy edited by Edward N. Zalta.

Landhuis, Esther. 2016. “Scientific Literature: Information Overload.” Nature 535: 457–458.

Mizrahi, Moti. 2019. “The Scientism Debate: A Battle for the Soul of Philosophy?” Social Epistemology Review and Reply Collective 8 (9): 1–13.

Peels, Rik. 2018. “A Conceptual Map of Scientism.” In Scientism: Prospects and Problems edited by Jeroen de Ridder, Rik Peels, and René van Woudenberg, 28–56. New York: Oxford University Press.

Plantinga, Alvin. 1974. The Nature of Necessity. New York: Oxford University Press.

Putnam, Hilary. 1975. Mathematics, Matter and Method. Cambridge: Cambridge University Press.

STM: International Association of Scientific, Technical and Medical Publishers. 2018.. The STM Report: An Overview of Scientific and Scholarly Publishing. The Hague: STM.

[1] There are now several international research centers devoted to research about research, e.g., the METRICS Center at Stanford, the Meta-research Center at Tilburg University in the Netherlands, and the London based Research on Research Institute.

[2] See Bourget & Chalmers (2014), De Cruz (2017, 2018a, 2018b) and also blog posts by Kieran Healy (2013) and Brian Weatherson (2017) for examples.

[3] Hence the failure of behaviorism as a research program in psychology and the subsequent reemergence of cognitive psychology; see Graham (2019) for some discussion.

[4] Thanks to Rens Vliegenthart for discussion about the methodology of Mizrahi’s paper. Research for this publication was made possible through the support of a grant from Templeton World Charity Foundation. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of Templeton World Charity Foundation.

Categories: Critical Replies

Tags: , , , , , , , , , ,

Leave a Reply