Ernest Gellner’s Use of the Social Sciences in Philosophy

1. Introduction

In the contemporary academic world, specialization is ever growing. The general trend is that people know more and more about less and less, and now and again, already tiny disciplines split into several new ones. This brings many epistemic benefits—for example, it obviously makes our knowledge more detailed— but it also has some costs attached to it. The most important of these is an opportunity cost: many chances to connect different topics are missed, because so few scholars have the necessary competence in all of the relevant disciplines.

One of the twentieth-century academics who most conspicuously ran counter to this trend was Ernest Gellner (1925-1995), who started out as a (renegade) philosopher, went on to join the sociology department at the London School of Economics and Political Science, and eventually wrote a doctoral dissertation in social anthropology (Gellner 1961) on Moroccan Berber society. He continued in a similar fashion for the rest of his career, becoming Professor of Philosophy with Special Reference to Sociology at two different departments at the London School of Economics—sociology (1962-1969) and philosophy (1969-1984)—before moving to Cambridge to take up the William Wyse Professorship of Social Anthropology. Gellner was the archetypical polymath, publishing on such diverse subjects as linguistic philosophy (Gellner 1959), psychoanalysis (Gellner 1985a), and the history of social anthropology (Gellner 1995). As would be expected in the age of specialization, Gellner was criticized for his broad interests and for his preference for staying at the general and abstract level rather than going into the miniscule details of the various debates he joined (he usually stayed in a particular debate for only a fairly short period of time before hurrying on to yet another one). ¹ The criticism was returned: to Gellner, specialists often failed to distinguish between technical details and things that really mattered. ²

Gellner’s thought is a great example of how fruitful breadth of interest and knowledge can be in the hands of a brilliant scholar. His philosophical knowledge was an important reason why he was such a great sociologist and anthropologist, ³ and conversely, his knowledge of the social sciences lay behind his highly original perspectives on philosophy. It is the latter influence that I am concerned with in this article. More exactly, I am interested in how Gellner used the social sciences (in particular, his areas of expertise: sociology, anthropology and the history of ideas) in philosophy. While it is fairly obvious that he did use them (this is pointed out, for example, by Jarvie and Agassi 1979), how he did it needs further elucidation.

2. A Philosopher or Not?

In formulating my problem as I do, I presuppose that Gellner was indeed a philosopher. Even though it might be seen as obvious that he was (after all, he was employed for many years as a philosopher and published a number of books that common usage would classify as philosophical ⁴ ), analytical philosophers have, as Ian Jarvie (1996, 521) points out, tended to see him as a “mere sociologist.” ⁵ Thus, his enraged attack on Oxford linguistic philosophy, Words and Things (1959), was seen as an anthropological or sociological work much like his work on Moroccan Berber society—the only difference being that Gellner was here studying the tribe of Oxford philosophers rather than Berber ones. Since most analytic philosophers think that a theory’s genesis has no relevance for its justification, this was a way of disqualifying Gellner’s criticism (Jarvie 1996, 521-22). Jarvie argues that this charge is quite unwarranted, claiming that while Words and Things does contain sociological or anthropological parts, this was not its only or even main content:

Four fifths of Gellner’s critique . . . was an entirely orthodox marshaling of arguments to show the incoherencies and inconsistencies of that school of thought. It displayed a detailed familiarity with the professional writings of all of the most influential of the British language philosophers of the time. (521)

Against the charge that Gellner was a “mere sociologist,” Jarvie claimed that Gellner was indeed a philosopher (albeit a philosopher with a sociological touch):

The temptation to think of what he does as sociological analysis of philosophy is in my view profoundly mistaken. It rests on the premiss that the narrow construction of philosophy which excludes his work is defensible. In fact his work undermines any defense of that narrow construction of philosophy and, indeed, proceeds on the basis of a quite different construction. Gellner’s own practice is to treat sociology as a specialized branch of philosophy. The important results achieved in this branch feed back and alter philosophy as a whole. (522)

Jarvie goes on claim that “his major books are almost all philosophical” (523). Given a wide interpretation of the notion of philosophy, this is patently true: most of Gellner’s work in the social sciences are fairly abstract and concern theoretical and methodological matters, such as the nature of nationalism (Gellner 1983) and the structure of historical development (Gellner 1988). Clearly, there is a use of the term philosophy according to which this sort of work should be labeled philosophical. However, Jarvie (1996) argues that Gellner was a philosopher not only in this wider sense but also in the narrower sense preferred by analytic philosophers. Thus, he insists that the claim that Gellner was a philosopher is “not based on some arbitrarily broad construction of philosophy as comprehensive [but] remains true even on the narrower traditional construction of philosophy held by my Oxford colleague” (523). ⁶ However, he then goes on to say that Gellner’s main philosophical interest was the role of science in the “emergence of the modern world” (523)—the escape from premodern Gemeinschaft to modern Gesellschaft. This is hardly a philosophical question in the narrow sense preferred by Jarvie’s Oxford colleague. Most analytical philosophers are not interested in such questions, ⁷ nor do they think of them as philosophical.

Consequently, Jarvie’s claim that Gellner was a philosopher does indeed rest on the premiss that the notion of philosophy is to be understood in a broad sense. This premiss is, however, not needed; as we will see, it is possible to show that Gellner was a philosopher in the narrower, “formal,” sense too. ⁸ In the following, we will look at a few examples establishing that Gellner was indeed a formal philosopher. These examples will also serve to show how he used his knowledge of the social sciences in formal philosophy.

3. Falsificationism as a Destroyer of Faiths

Since Gellner’s philosophical output was very rich (even given the restricted notion of philosophy used here), my discussion will be very short of comprehensive. Limited scope forces me to focus on a few examples that I deem to be especially interesting and enlightening for the present purposes. While I am sure that my discussion leaves out many aspects of this question that deserve to be commented on, I hope that my general picture of how Gellner used the social sciences in philosophy is roughly correct.

My first example concerns Gellner’s discussion of the difference between Karl Popper’s falsificationism (Popper [1934] 1959) and the logical positivists’ verificationism. According to the standard view, espoused by Popper and the verificationists, the debate between them mainly concerned various technical philosophical problems, such as the problem of induction or the logic of valid scientific inference. To get a sense of the nature of these problems and the debaters’ theories of how to solve them, let us look at falsificationism’s proposed solution to the former problem. Ever since Hume pointed out that inductive arguments—which have finite numbers of observations as premisses and general propositions such as “All As are Bs” as conclusions—are not deductively valid, philosophers—in particular, verificationists such as Carnap (1950), according to whom science is based on induction—have discussed the nature of inductive validity. Popper’s proposed solution was characteristically bold: in fact, inductive arguments are not valid, but neither do we need them in science! For, as Popper saw it, scientists do not aim at verification of general propositions but at falsification of them while provisionally accepting as-yet-unfalsified propositions. And, according to Popper, falsification of general propositions is, contrary to verification of them, not logically dubious (an issue that of course has been hotly debated).

But as Gellner saw it, these technical differences were not the heart of the matter. What really set falsificationism apart from verificationism was its superior ability to reveal the faithlike character of theories such as Freudianism and Marxism:

What really makes the criterion of falsifiability so powerful is this: if you insist that a believer specifies the conditions in which his faith would cease to be true, you implicitly force him to conceive a world in which his faith is sub judicie, at the mercy of some “fact” or other. But this is precisely what faiths, total outlooks, systematically avoid and evade. They fill out the world of their adherents, the world they in a way create, and they interpret the processes of cognition in such a way that all verdicts must, in the end, be returned in their favour. Note that they have little to fear from a requirement that they be “verifiable”: generally speaking, they pervade the world they create so completely that verifications abound—here a verification, there a verification, everywhere a verification. Ironically . . . the verification principle penalizes only the honest faith, which segregates and candidly admits the “transcendent,” unverifiable elements within it, like the honest traveller who admits to all his purchases that are liable to customs duty. Those who systematically mix up the transcendent with their other luggage have little to fear from a customs examination by the principle of verification. (Gellner 1974b, 176)

To explain Gellner’s argument, we need to make a slight detour at this point. For Gellner, falsificationism had much more in common with classical empiricism and positivism than both Popper himself and his critics thought (see, e.g., Gellner 1985b, 4-67); indeed, it was but the last and most sophisticated of a long line of empiricist attacks on faithlike theories. Thus, consider Gellner’s view of premodern religion and the empiricist critique of it. What characterizes such religions is, according to Gellner, that the boundary between the empirical and the transcendental is not recognized: “concepts habitually and constantly dart across this boundary, and the boundary itself is barely perceived” (Gellner 1974b, 163). They are simultaneously given both an empirical and a transcendental interpretation. Thanks to this quality, they give the believers substance (provided by the empirical interpretation) as well as certainty (provided by the transcendental interpretation). The gist of the classical empiricist critique of premodern religion was a relentless insistence that this boundary existed and should be respected. Theories purporting to describe the nature of the material or immanent world needed, first, to be clearly separated from those that did not and, second, to be based on empirical evidence. The most succinct encapsulation of this attitude is of course Hume’s:

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. (Hume [1739] 2004, 107) ⁹

Of course, the Humean attitude conquered, and faiths would never again dominate the Western world. But they did not die out. Religion was gradually transformed in the face of empiricist criticism: doctrines easily refuted by Hume’s dictum were replaced by ones that were not. Also, new “faiths” even more compatible with the modern worldview—such as Marxism and Freudianism—rose to prominence. Traditional empiricist theories such as verificationism—which were effective enough against earlier faiths, such as premodern Christianity—failed to hurt these theories, since they, as Gellner notes, saw verifications of their theories everywhere. It took a new empiricist weapon, falsificationism, to hurt these evolved faiths. ¹⁰

What made it possible for falsificationism to hurt them is this. Falsificationism is built on the observation that if a piece of evidenceE empirically confirms or verifies a hypothesis H, then not-E by necessity disconfirms or falsifies H. ¹¹ Thus, unless some potential states of affairs falsified or at least disconfirmed H, it could not have any empirical support nor, indeed, any empirical content whatsoever. Therefore, examples of such potential falsifiers had to be provided for the theory to count as empirical or scientific. But, providing such potential falsifiers “disenchants”—to use another one of Gellner’s favorite terms (borrowed from Weber, of course)—the faith by forcing its adherents to confront the possibility that it might be false. It loses its privileged position—it no longer “fills out” the world—and becomes exposed to critical assessment just like other theories.

Gellner’s reinterpretation of Popper points to an important aspect of the latter’s thought. Popper was, of course, unusual in that he was both a traditional formal philosopher and a social philosopher, with a fervent interest in politics and the social sciences. To him, these topics were all of one cloth and neither could nor should be separated. There are obvious affinities between Popper and Gellner on this point, but there is also a significant difference: the connection between formal and social philosophy was much more direct and unproblematic for Popper than it was for Gellner. According to Popper, social and political problems were often caused by flawed formal philosophies, such as “essentialism” and “historicism” (Popper [1945] 1950, 1957). It was up to the formal philosophers to solve these problems so that social and political development could continue unimpeded. Moreover, the task was not to just give a sketch of a solution but to conclusively solve the problems using all the technical machinery of modern philosophy. Thus, falsificationism was supposed to save the world from the social and political evils that Freudianism and Marxism brought upon it by solving highly technical problems such as the nature of valid scientific inference. This put it in a very strong position rhetorically: it gave it both political and social relevance on the one hand and respect as technical philosophy on the other. Most philosophical theories have either the first or the second quality: relevance is often bought at the expense of exactitude and vice versa, but falsificationism seems to have both of them.

Gellner had a somewhat different view of the relationship between formal philosophy and the social and political development. To him, philosophy was not and could not be an exact discipline. ¹² He much liked what Kripke called philosophical “pictures” (1980, 93) and did not care much for the subsequent modifications and qualifications of such pictures or general theories (see, e.g., Gellner 1974b, 162). ¹³ Developments in formal philosophy did have important social and political consequences, but the crucial factor was the general imperatives that the pictures expressed—”Make your theories falsifiable!”, “Maximize happiness!” ¹⁴ and so on—rather than the subsequent fine-tuning of the pictures. Moreover, the supposed exactness of the sophisticated versions of the original pictures was usually illusory.

Thus, to Gellner, falsificationism was really a destroyer of faiths: the fuss about the exact logical form of scientific inference, the problem of induction, and other technical problems of philosophy was just rhetoric. It was falsificationism’s general imperative and its ability to destroy faiths that actually mattered. This amounted to a quite substantial reinterpretation of Popper’s thought – a reinterpretation that squared with Gellner’s own interests. In particular, it was his preoccupation with belief systems, especially closed belief systems or ideologies (Gellner 1978), that disposed him to interpret Popper’s falsificationism as a powerful tool to disenchant and ultimately destroy such ideologies, rather than as a solution to various technical philosophical problems. We will see that the tendency of Gellner’s to reinterpret problems so that they fit his own interests recur in his writings, and we will sometimes have reasons to question its tenability.

4. Are We Prisoners of Our Language Games?

My second example concerns the question whether we can break free from our “language games” (Wittgenstein) or “frameworks” (Popper). As we shall see, Gellner struck a middle path between (the later) Wittgenstein, who believed it to be impossible, and Popper, who thought it was very easy.

Let us first look at his view of Wittgenstein’s theory of language games, which he commented extensively on already in Words and Things (Gellner 1959) and repeatedly returned to (e.g., in his very last book; Gellner 1998). Even though Gellner was very critical of Wittgenstein, there were certain similarities between their philosophies. Gellner thought that Wittgenstein was right to emphasize the need for philosophers to observe how our language games or “forms of life”—that is, cultures—actually work. ¹⁵ Philosophical problems arise in concrete social contexts, and if those social contexts are neglected, as they usually are by analytic philosophers, we tend to end up with an impoverished philosophy.

Thus, Gellner agreed with Wittgenstein that we need more descriptions of our language games or forms of life in philosophy. But, he disagreed with Wittgenstein’s view of how we should make those descriptions and observations. There are, of course, whole disciplines whose very essence is to observe and describe forms of life empirically—in particular, sociology and anthropology—and to Gellner, it was self-evident that philosophy should make use of the knowledge obtained by them. Wittgenstein and the other linguistic philosophers did not do that, however, but rather tried to base philosophy on their own peculiar “armchair observations” (which sounds like and probably is a contradiction in terms). Gellner’s (1974a, 33) verdict on this method is clear:

The way forward does not lie in amateur and comically timeless linguistic sociology which takes “forms of life” for granted (and this is what philosophy has been recently), but in the systematic study of forms of life which does not take them for granted at all.

In spite of all their talk of “observation” and “description,” linguistic philosophers were very traditional philosophers in the (very important) sense that their philosophy was done from the armchair. They did not take their supposed “revolution in philosophy” to its logical conclusion—namely, that philosophy must be empirically grounded and tightly integrated with the social sciences. ¹⁶

Gellner argued that this armchair methodology gave rise to a conception of language games or forms of life as being much more consistent, clearly delimited, and mutually exclusive than they actually are. In fact, one could argue that just as previous philosophers were, according to Wittgenstein, captivated by a simplified notion or picture of the nature of language (i.e., that all sorts of language use are referential), he himself was captivated by a simplified notion of the nature of language games. We use such simplified notions because we do not need (nor have time to work out) more nuanced ones in everyday life. However, in science and in philosophy, we do need to replace these simplified notions by more accurate ones. Some simplified notions—such as the picture of language as purely referential—can be dissolved by Wittgenstein’s preferred method of pure armchair reflection, but others—such as the view of language games as more homogeneous and clearly separated than they actually are—can be dissolved only by empirical (i.e., anthropological and sociological) research.

This misconception of the nature of our language games gives rise to a further misconception—namely, the idea that we can infer how we should use our concepts from observations of how they actually work. As Wittgenstein saw it, our language games already provide us with clear and unambiguous rules, and thus there is no need to systematically reflect over what rules we should use (i.e., to philosophize, as most people use that term). Instead, we should follow those rules that already are in place. Indeed, it is not even possible to make such reflections, since we cannot transcend our language games. It is the illusion that such reflections—such criticisms of our language games—are possible that engenders philosophy. Thus, philosophy consists of pseudo-reflections that arise when language “goes on holiday” (Wittgenstein [1953] 2009, 23) – when we start making theoretical speculations on our concepts rather than putting them to concrete use.

Gellner has several arguments against this view (Gellner 1959, 157-58), but the one that concerns me here is his view that since we are more often than not playing several overlapping games at once, our language games do not provide us with any clear guidance. We are not given unambiguous rules but rather a conceptual mess in which we fail to orient ourselves (157). The philosopher’s job is to enable us to orient ourselves in this mess and, ideally, to bring some order to it. ¹⁷ This means that we are not in fact prisoners of our language games and that criticism of them is not pathological. Indeed, the fact that it is possible to break free from prevailing language games—and that people such as Descartes and Luther did—is “probably the single most important fact about the intellectual life of mankind” (Gellner 1985b, 185).

It would perhaps seem that the preceding argument puts Gellner very close to Popper, who of course was the arch enemy of the “myth of the framework,” as he termed it (i.e., the view that we are prisoners of our frameworks or language games). However, Gellner thought that Popper went too far when he claimed that “if we try, we can break out of our framework at any time” (Popper 1970, 56, quoted in Gellner 1974b, 182). “At any time?” Gellner asks, rhetorically. No, “it was not so easy” (Gellner 1974b, 182). Our language games or cultures do constrain our thoughts, making a radical break with tradition a substantial feat. Moreover, some of them—so-called closed societies—constrain us more than others—open ones. It takes a Descartes or a Luther to challenge the entrenched conceptual order of a closed society; not everyone has the necessary imagination and courage to pull that off.

What I find most interesting here is not the fact that Wittgenstein’s extreme fatalism and Popper’s extreme libertarianism are wrong, whereas Gellner’s more balanced position is right (though I do not doubt that this is the case). ¹⁸ Instead, my central concern is with Gellner’s method and how it differed from the method that Wittgenstein and Popper actually shared (though they tended to disagree on most other philosophical matters). Both Popper and Wittgenstein seemed to think that the question whether we can rebel against the prevailing conceptual order can be answered from the armchair, with no, or a minimum of, references to the plethora of cultures that have existed throughout history. This is, of course, in line with the aprioristic methodology that is dominant in analytic philosophy. ¹⁹ An important consequence of this aprioristic methodology is a tendency to be drawn toward extreme positions. The notion that we are always free could be defended from the armchair, as could the notion that we are never free, but we cannot establish that we are free under (only) some specified circumstances without reference to empirical facts.

The question “Are we prisoners of our language games?” is obviously quite vague and can be interpreted in different ways. Some of these interpretations will make Wittgenstein right by definition. For example, a defender of Wittgenstein’s position could argue that Luther did not break out of the Catholic language game but rather just carried out reforms within it. Gellner (1973, 50-72) rejects this move when he comments on Peter Winch’s (1958) interpretation of Wittgenstein. ²⁰ Similarly, we could interpret the question so that Popper comes out as winning. However, we are obviously not interested in such idiosyncratic and special versions of the question but rather in its “normal” or “natural” version. Given the question’s vagueness, we are of course not completely clear over what this natural version or reading is, but even so, we are sufficiently clear over it to be able to say with some confidence that Gellner is right: when the words that the question is composed of are used with their standard meanings, it is possible, but often hard, to break free from our language games.

Both the Popperians and the Wittgensteinians obviously wanted their respective theories to answer the natural version of the question rather than some special philosophical version of it. They intended their theories to be substantive—that is, to have empirical content—rather than merely conceptual. ²¹ The reason why their theories seem more plausible than they are is that their statuses are not fixed: they slide back and forth between a substantive (or natural) and a conceptual (or philosophical) version. ²² Once we disallow such sliding, as we of course should, and settle for the natural version of the question, it becomes obvious that we should adopt Gellner’s empirically oriented method rather than Popper’s and Wittgenstein’s armchair method.

5. The Positivist Theory of Meaning

My last example concerns Gellner’s view of the logical positivists’ theory of meaning, which said that the meaning of a statement is its method of verification. On the face of it, it seems like this theory is a descriptive theory about how we do speak, but according to Gellner (1985b, 30-31), it is rather a normative theory about how we should speak:

The only thing which in effect they could mean, plausibly and in harmony with their other principles, was this: the definition circumscribed, not the de facto custom of any one or every linguistic community, but the limits of the kind of use of speech which deserves respect and commendation. It was a definition not of meaningful speech, but of commendable, good speech. Their verificationism was a covert piece of ethics.

There is much to be said in favor of Gellner’s interpretation. Lots of statements that clearly are meaningful, such as “the Ancien Régime was morally repellent,” come out as meaningless according to the positivists’ criterion. This means that if the theory were to be interpreted as descriptive, it would stumble on the first hurdle. Combining this with the notion that it is scarcely credible that philosophers of the logical positivists’ class would have put forward a theory that is obviously false, it follows that they must have meant something else. Moreover, there are numerous positive reasons for interpreting their theory as normative. For example, the verification principle was their main weapon in the intellectual battle they fought against (what they perceived as) confused and antiscientific philosophers in Germany and Austria (see, e.g., Carnap 1931).

Even so, there are reasons to believe that Gellner’s interpretation is, if not wrong, then at least incomplete. For, even though it is true that the logical positivists’ implicitly treated their theory as normative, at least most of the time, the fact that it was formulated as a descriptive theory did have consequences. Most importantly, their chief critic Quine went for the literal interpretation (as analytical philosophers are inclined to do) and hence took the theory to be a descriptive one in his “Two Dogmas of Empiricism” (1951). Consider, for example, his criticism of Carnap’s attempt to clarify the notion of analyticity in terms of semantical rules:

Appeal to hypothetical languages of an artificially simple kind could conceivably be useful in clarifying analyticity, if the mental or behavioral or cultural factors relevant to analyticity— whatever they may be—were somehow sketched into the simplified model. But a model which takes analyticity merely as an irreducible character is unlikely to throw light on the problem of explicating analyticity. (34)

If Carnap’s goal would have been to describe and explain how language actually works, then not to include “mental or behavioral or cultural factors” would have been a serious mistake. But as we saw, this very fact makes it likely that Quine’s interpretation is uncharitable and incomplete. It would thus seem that Carnap is using these artificial languages not to understand how natural language actually works but rather as a sketch of a reformed, better language.

Still, most philosophers took Quine’s criticism to be very damaging to Carnap’s and the other logical positivists’ project. After all, the positivist theory of meaning and analyticity purported to be descriptive, and thus it should be judged as such, they thought. Hence, the logical positivists’ choice to cast their theory as descriptive, even though it in fact was normative, was not insignificant but had important historical consequences. Also, Carnap was not very clear on this point and occasionally seemed to think that his theory adequately described how (parts of) scientific language actually functions (e.g., Carnap 1934). Thus, there is a sense in which he after all “meant” what he literally said. The question is not as unambiguous as Gellner presents it.

I do not think that Gellner would have denied this. Indeed, he had some very subtle things to say about the interpretation of statements that seem to blatantly contradict the utterer’s other beliefs and statements. ²³ Gellner’s texts are more often than not brimful with ideas, and presumably, he had to skip these qualifications to make room for other points that he deemed more relevant in the context in question.

Nevertheless I think that this omission of Gellner’s can teach us something about his way of thinking and writing. He obviously had a voracious appetite for knowledge (tellingly, he seemed very much preoccupied with Faust) and again and again turned to new subjects that he felt not to have grasped sufficiently. Most of all, he craved for overviews: general pictures of various areas (the history of philosophy, the nature of nationalism, the structure of historical development, etc.). This preoccupation with foundational questions stemmed mainly from a very vivid sense of the precariousness of our epistemic situation: to him, we were more often than not “lost, say on a long walk in bad weather in uninhabited country” (Gellner 1974b, 10) both in philosophy and in the social sciences. His aim was not to create full-blown systems but rather to allow us to regain our orientation. As a natural consequence, his theories usually lacked qualifications. Even though qualifications and refinements may take our theories closer to the truth, they usually make it harder to get an immediate grasp of the latter. Qualifications have their place at a later stage, but when the job is to find us a way through the fog, they must be omitted.

The quest to help us regain our orientation is in many ways a commendable one. We do need overviewing theories that are explicitly stated and argued for (a point that Gellner makes in his Plough, Sword and Book; 1988, 11-15), and Gellner gave us such theories in many areas. However, the lack of detail and refinements sometimes led to oversimplifications. Thus, writing that the logical positivists meant that the verification principle was purely normative distorts the role that it played in their system and in the wider philosophical discussion. Had it been put forward as an overt rather than covert piece of ethics (see above), Quine would not have attacked it for its flaws as a descriptive theory, nor would it presumably have been as widely accepted in the first place. Hence, Gellner’s attempts to understand the essence of various philosophical positions—in this case, logical positivism—sometimes leads to a neglect of complicating factors. Reading Gellner, you need to be attentive: he is not one of those writers who give you everything; rather, you need to fill in the blanks yourself. Yet, this is no doubt an important reason why he is so enjoyable to read: his style of writing invites you to think, to make associations of your own, and to assess whether the breathtakingly bold assertions that you meet on page after page really are true.

6. Conclusions

We have studied three examples of how Gellner used his knowledge of the social sciences to construct highly original theories of central problems of formal philosophy. It might be argued that some of these theories are not first-order theories about philosophical problems but rather metatheories about the nature of other philosophers’ first-order theories (such as falsificationism and the verificationist theory of meaning). However, this distinction is hardly tenable in this case, since Gellner’s reinterpretations of these theories had a lot of philosophical content; for better or worse, they were not impartial “pure descriptions.” Hence, we can conclude that Gellner really was a philosopher not only in the broader sense but also in the narrower, formal sense. In fact, his philosophy was, to my mind, at least as brilliant as his work in the social sciences. Due to the previously mentioned specialization of the academia, he has, though, not got the recognition he deserves for it. Outside of formal philosophy, few are capable to appreciate the ingenuity of his reinterpretations of philosophical theories and problems, whereas formal philosophy itself is dominated by analytic philosophers, who do not care for Gellner’s broad-brush style but prefer their own method of looking at each philosophical problem in close detail and in isolation from the social context in which they arose.

But let us turn to my main question: how did Gellner use his knowledge of the social sciences in philosophy? Even though it is, as stated, not possible to construct a comprehensive theory about this on the basis of the few examples that I have discussed, I think that some conclusions can be drawn. In the first and third examples, Gellner uses his knowledge of the social sciences to make bold “translations,” or reinterpretations, of various philosophical problems and theories. Thus, falsificationism, usually seen as a narrowly logical doctrine, was translated into a destroyer of faiths of major relevance for the social sciences and indeed politics. Likewise, verificationism was translated from a descriptive theory to a normative one. Gellner’s treatment of the debate over whether we are trapped by our “language games” is a little bit different: in this case, the participants already viewed the problem as relevant for our understanding of historical and social reality, and Gellner just pointed out that given this interpretation, the problem had to be solved by reference to relevant empirical facts rather than by armchair philosophy. Thus, in this case, Gellner left the problem as it was but argued that it had to be approached by different methods.

Gellner’s argument in the second example seems to me correct and unproblematic, so let me concentrate on his translations of philosophical problems and theories. I have taken the notion from Gellner (1974b, 31) himself, who calls his interpretation of the epistemological tradition as normative rather than descriptive a “translation.” However, Gellner seldom reflected on his philosophy as a whole and did not put forward any general metaphilosophical theory concerning why such translations are needed and what criteria they should fulfill. ²⁴ In any case, it is clear that this kind of translation is a very important aspect of Gellner’s philosophy. The translations do have many attractions; for example, they often make seemingly scholastic problems highly relevant. Also, they have the potential to save us from making uncharitable interpretations of philosophical theories. ²⁵ Overall, I highly recommend this methodology, but it does carry some risks with it. There is a thin line between the original and the idiosyncratic, and some of Gellner’s interpretations strike me as having crossed that line. These interpretations are too far removed from what the texts in question actually say and are overmuch influenced by his own interests and concerns. This is particularly true of his interpretation of Descartes’s theory of how we can obtain secure knowledge in spite of the fact that we may be tricked by an evil Daemon. ²⁶ Descartes’s speculations have seemed tortuous and implausible to many twentieth-century philosophers, but to Gellner, this criticism is naïve and ahistorical. Instead, he thought that Descartes should be praised as one of history’s most important epistemic heroes who, first, dared to admit that we are “lost”—that we inhabit an epistemically insecure world—and need to regain our orientation and, second, tried to give a systematic theory of how we could do this. He saw Descartes as personifying a new cognitive ethos—a rationalistic ethos that demands reasons for all beliefs—that has transformed the world. This observation is, I think, more or less correct, and he is right in attacking the critics of Descartes for not seeing this (see, e.g., Gellner 1979, 148-63), but it should not blind us to the scholastic aspects of Descartes’s philosophy.

Gellner’s interpretations in my main examples are not as idiosyncratic, but even so, his tendency to bend philosophical problems and doctrines so that they fit his own worldview is obvious. The logical differences between falsificationism and verificationism are much more important than Gellner would have it. Likewise, his reinterpretation of the verification theory of meaning distorts the true nature of that theory so that it better fits into Gellner’s view of empiricist theories as “ethics of cognition” (see Gellner 1979, 164-81).

These objections are, however, minor and do not blind me to the subtlety of Gellner’s philosophy. While many of his reinterpretations of technical philosophical problems are highly interesting in themselves, the most exciting aspect of his philosophy is the general translation method that he developed. There is a seed of a whole metaphilosophical program here, which still waits to be developed in detail.