“Problems Should Completely Disappear”

The claim that there are no anthropological problems might sound like an improbable proposition, but there are coherent and defensible reasons for which people might declare explicitly or imply such an answer. Take, for instance, the commonplace that writers should be in the business of “showing,” not “telling,” and its implication that description trumps explanation or interpretation. This is occasionally applied explicitly to ethnography (e.g., Gullion 2016), but almost any understanding of ethnographic writing will rely, to some degree or another, on the importance of narrative and imagination. “Imagine yourself suddenly set down …” is an attempt by Malinowski to bring his readers along with him, to evoke, not just to state or aver a set of ethnographic facts. Despite Malinowski’s stress on the importance of collecting and presenting a “scientific” array of statistical and observational data in ethnographic writing, he was also no stranger to the importance of engaging the reader’s imagination in order that they “conceptualize … what the text could not present in full” (Thornton 1985, 8), hence his notion of “imponderabilia.” In a slightly different vein, Evans-Pritchard suggested that an anthropologist’s “theoretical conclusions” should be “found to be implicit in an exact and detailed description” (1973, 3). Though that may have a distinctly Latourian ring to a twenty-first century ear, it follows a series of arguments about the necessity of possessing a general theoretical framework within which such a description would make sense: “One cannot study anything without a theory about its nature” (2). The notion that ethnographic writing must involve some degree of “showing” is of course not the same as the claim that “showing” exhausts what ethnographic writing should be doing. That latter claim implies that objects of ethnographic writing require no further elucidation than to be “shown”; they are in themselves revealing of whatever it is that is to be revealed. They are not problems requiring solutions. They need only be properly depicted or described for their nature to be understood.

A related but much more philosophically rigorous justification of this sort of position returns us to Popper’s adversary in his argument for the existence of problems in philosophy, for it is in Wittgenstein that it finds its greatest exponent. Wittgenstein’s views need no lengthy exposition, and I am far from capable of doing them justice here, so a brief summary must suffice:

It was true to say that our considerations could not be scientific ones. It was not of any possible interest to us to find out empirically “that, contrary to our preconceived ideas, it is possible to think such-and-such”—whatever that may mean … and we may not advance any kind of theory. There must not be anything hypothetical in our considerations. We must do away with all explanation, and description alone must take its place. And this description gets its light … from philosophical problems. These are, of course, not empirical problems; they are solved, rather, by looking into the workings of our language, and that in such a way as to make us recognise those workings: in despite of an urge to misunderstand them. The problems are solved, not by giving new information, but by arranging what we have always known. (PI, 109)

What are the “scientific considerations” that Wittgenstein thinks philosophy should avoid? They are, broadly speaking, general and reductive explanations, which are “the real source of metaphysics, and lead … the philosopher into complete darkness” (The Blue and Brown Books, 18). Famously, for example, we do not grasp the meaning of a word by possessing a mental picture of a general idea. We do not know how to point to a “leaf” because we have a general idea of a “leaf” (18), or to a “yellow ball” because we have a picture of “yellow”: “To see that this is not necessary, remember that I could have given … the order, ‘imagine a yellow patch.’ Would you still be inclined to assume that he first imagines a yellow patch just understanding my order, and then imagines a yellow patch to match the first!” (11–12). So here we have a vision of problems that are absolutely not amenable to empirical resolution. Nothing—according to Wittgenstein, in any case—can answer as a matter of fact the question of what goes on in someone’s mind when ordered to point out the yellow ball, a thing that we call “understanding.” What resolves this puzzle instead is linguistic clarity. When we use the word understanding, we often imagine it to exist as a full-fledged mental state, complete with mental picture of the general idea of a phenomenon. But Wittgenstein shows us—note the showing, rather than explicit telling—that this makes no sense of our ordinary use of the word understand: nobody would require that possession of a mental picture of a yellow patch be a prior requirement to correctly fulfilling the order to conjure up a mental picture of a yellow patch. And in fact, in practice, it is the correct fulfillment of such an order that we take to constitute “understanding” in everyday life—we say the man has “understood” the order if he picks out the yellow ball, not (usually) by checking to see if he has the right mental picture.

It is not difficult to imagine anthropological equivalents of this vision in which terms require rearrangement and clarification for the puzzle to dissolve. Indeed, almost any anthropological argument that relies on some version of the “category mistake” problem (itself a Wittgensteinian notion that comes from Gilbert Ryle [1949]) does this to some degree or another. When Hawaiians killed Captain Cook, were they killing a man, a god, or a chief (Obeyesekere 1992; Sahlins 1985, 1995)? We might put that question on one end of the scale, given the amount of empirical data mustered in support of the various positions. Is paternity a physical, biological relationship of which one can be ignorant, or a conceptual, social relationship that expresses cosmological meaning (Delaney 1986; Leach 1966; Spiro 1968)? This latter is a “purer” Wittgensteinian position, in which it is clearly stated that the “problem” of virgin births is a product of misconstrued meaning and the misposing of questions (Delaney 1986, 494). Are Ifá oracular pronouncements meant to be representational claims about the world, or “inventive definitions” that transform their objects (Holbraad 2009, 2012)? This last example is perhaps as clear an example as is to be found in anthropology, and it has been accompanied by a whole theoretical movement that explicitly figures anthropology’s main task as one of reconfiguring concepts in order to render “alterity” sensical (Holbraad and Pedersen 2017).

“His Look Was Fleeting, and He Saw Very Little”

It is not my intention to dispute any of these particular arguments, and the broader “category mistake” framework is probably a fundamental and unshakable form of anthropological reasoning, though it is not always uncontested. Evans-Pritchard’s Azande is in a sense a response to the category mistake–type argument of Lucien Lévy-Bruhl on primitive mentalities. How can otherwise rational people believe in witchcraft? The Levy-Bruhlian answer would be to find this question misposed, assuming as it does that “rational” and “magical” thought are part of the same set. Evans-Pritchard’s (1937) argument and ethnography demonstrate instead just how perfectly possible it is for otherwise “rational” people to attribute causality, in a certain specific sense, to witchcraft.

Similarly, Wittgenstein’s views have not been without their critics in philosophy. Popper aside, Ernest Gellner wrote a famously scathing critique in Words and Things (1960; see also Gellner 1988); more relevant to present purposes is The Grasshopper, by Canadian philosopher Bernard Suits, who framed his argument as follows: “ ‘Don’t say,’ Wittgenstein admonishes us, ‘there must be something common or they would not be called “games”—but look and see whether there is anything common to all.’ This is unexceptionable advice. Unfortunately, Wittgenstein himself did not follow it. He looked, to be sure, but because he had decided beforehand that games are indefinable, his look was fleeting, and he saw very little” (1978, x).

The rest of The Grasshopper is a highly engaging series of logical demonstrations to the effect that games are indeed definable, but I raise Suits’s critique because of what it suggests about Wittgenstein’s method and object. I have already noted that the PI is not intended to be explanatory, but illustrative or elucidatory. What Suits makes clear is that the thing on which it is performing that operation, whatever one calls it, is not the category “games,” and certainly not any actually existing game. An alien wishing to learn about, to have something explained to them of, or elucidated on, football or boxing would do better to pick up Eduardo Archetti (1999) or Loïc Wacquant (2004), and even one simply wishing to learn about the category “games” would probably learn more from Suits than from Wittgenstein.

As far as Wittgenstein goes, this is unsurprising and unobjectionable, since clearly the PI is not meant to be a book about games, but about language (or language-games, perhaps, but the point remains). His anti-explanatory method, in other words, is applied to a particular kind of object. He was not, as Suits suggests, actually “looking [at] and seeing” “games” but rather looking at language. As Wittgenstein himself puts it, “Philosophy just puts everything before us, and neither explains nor deduces anything. Since everything lies open to view there is nothing to explain” (PI, 126). It concerns, in other words, things—language, above all else—that “lie open to view”: “The problems are solved, not by giving new information, but by arranging what we have always known” (109). In other words, his method demands that he can only be “looking [at] and seeing” something he already knows—like language—not something to which new information would be relevant.

While that view of philosophy may have much to recommend it, transposed to anthropology it becomes much more difficult to maintain consistently. Philosophy as an activity may well be the resolution of puzzles to which we already know the answer, in which “everything lies open to view,” but it is harder to imagine that all of the objects of anthropology are known to us already in this way, and that in no cases are new empirical data helpful in resolving our problems. The classic vision of the ethnographic puzzle had an ethnographic (i.e., empirical) solution, as in Audrey Richard’s various examples of concrete resolutions of the conflict between matriliny and patriarchy. Sometimes “ethnographic puzzles” may well turn out to be Wittgensteinian puzzles, of the sort requiring dissolution rather than solution. But often they concern situations in which everything does not lie open to view, and are thus much more like the “problems” Popper hoped to identify for philosophy: they require explanation.

In the remainder of this chapter, I am going to suggest that these two different perspectives on the character of anthropological problems have very different consequences. I do not suggest that all anthropological problems are problems requiring explanations. But neither do I think it sustainable to take the position that no anthropological problems have this character, and that position is a logical consequence of the notion that anthropology should not be in the business of explaining. In other words, I think that to inquire about and to make claims for or against “explanation” is not just a question of approach, method, or writing but also necessarily a question of objects. Whatever we are doing (explanation, interpretation, understanding, description, analysis), are we doing it to things we already know, or things about which we wish to know more?

To explain my position more clearly, I proceed to treat an example of what we might think of as an ethnographic puzzle, one focused on a topic that is quite closely related to explanation—namely, “definition.” Definition is an especially happy example to discuss for two reasons, beyond its kinship with “explanation” as an allegedly “reductive” activity.

First, it was problems of definition that led anthropologists such as Edmund Leach (1961) and, building on the same ideas, Needham (1971, 1975) to problematize in turn the possibility of explanatory generalization, although neither of them, in fact, wished us to abandon generalization altogether (Leach 1961, 1; Needham 1975, 365). Leach famously, in “Rethinking Anthropology,” described the categorizations that led anthropologists to “problems” such as that of matriliny as instances of “butterfly-collecting,” declaring that the “problems” originated in the arbitrary categories (“matrilineal society”), not in the world (1961, 2–3). Elsewhere Leach (1984, 17) acknowledged the influence of the later Wittgenstein on his vision of anthropology, but this is much more pronounced and explicit in Needham, who, making a very similar argument a decade later than Leach, in “Remarks on the Analysis of Kinship and Marriage” (1971, 2), cites Leach and Wittgenstein in the same breath as inspirations.

Second, we have, once more, Wittgenstein in general and Needham in particular to thank for the notion that it is not only our anthropological categories that escape definition but also most ethnographic ones we are likely to encounter. Needham’s article on the Wittgensteinian notion of “polythetic categories” is famous, like its predecessor, for the argument that the class of societies we once termed “matrilineal” or “patrilocal” in fact contain no single shared predicate that justifies their definition as a set (1975, 365). Worse, even more general categories such as “descent,” “marriage,” or “kinship” are equally unable to pick out substitutable objects, or equivalent meanings. For these reasons Needham argued that we treat such categories as “polythetic,” or as what Wittgenstein would call “family resemblance terms,” like the category “games.” Polythetic categories and family resemblance terms are not united by any essential quality or predicate that all members of the category share but are instead woven together like a rope, in which different fibers create the stability of the entity, rather than any single one (350). But Needham not only believed that most of our analytical terminology was polythetic in nature; he also believed—unsurprisingly given the influence of Wittgenstein—that ordinary language was too. That is, not just “our” ordinary language but any instance of ordinary language that an anthropologist was likely to encounter, no matter what the indigenous perspective on such a matter might be—“alien concepts” are polythetic, “in a fashion that is similarly unrecognized by those whose modes of thought we want to comprehend” (367).

My intention is not to indict Leach or Needham, who both actually defended explanatory generalization of their own preferred relational form (Needham 1975, 365), but only to indicate that in treating a problem of definition, I am treating a problem central to the question of anthropological problems, as well as one that is also, in some sense, a quintessential example of what a nonproblem looks like to a certain sort of anthropology. That is, if you take the position that all concepts and categories are polythetic in nature (i.e., that they share no single quality or predicate), then any problem of definition is—by definition, as it were—not in fact a problem. If you know already that definition is impossible for you as well as for your interlocutors, then you need not worry about explaining—rather than describing—any given instance of it beyond stating that fact.

To illustrate how this may be unsatisfactory, I will describe a problem of definition I have encountered in the course of recent fieldwork on fascism in Italy. Before doing so, however, I want to do a little more to characterize the Wittgensteinian approach to definitions in anthropology, drawing on Needham in particular.

The Puzzle of Definitions

Any specific practice of definition that may appear to give rise to a problem, as I have just suggested, has a very easy answer if you take the position that going in search of definitions is necessarily a fruitless endeavor. Or rather, such a problem is dissolved because what is in fact going on is clearly something other than a practice of definition as we might be wont to imagine it—what is going instead are moves in a game, say, or perhaps the manipulation of terminology to suit a particular set of aims. In other words, we may know already, having read Needham or Wittgenstein, that defining words or categories in the way we imagine definition to work is impossible because words or categories do not consistently pick out the same objects or sorts of object in the world, being rather tools or devices with which to intervene in that world, the objects in which, in any case, do not possess essential qualities or properties that could be picked out by words. Knowing this, we may concern ourselves instead with describing instances of definition, with what they do, with their context, their tone and inflection, with the intentions we may or may not be able to read behind them, et cetera.

I hasten to add at this point that arguing for explanation is not the same as arguing against description. Clearly such description is a necessary condition for an anthropological account of whatever one wishes to call the activity of trying to define a concept. I am going to suggest, though, that it may not always be a sufficient one.

Consider the précis of the kind of account I have just provided. Here we have what some (Popper and Gellner, for instance) would call a doctrine, an explanation, a theory, or a “telling,” and what others might want to call an illustration, an elucidation, or a “showing,” but which in either case is performing an operation on “definition” qua definition as a philosophical question. The nature of that explanation/elucidation is to tell/show that definition is impossible. Then you have a descriptive activity that shows some particular (pseudo-)definition in action as composed of all of the contextual factors that really matter to it.

We can see this sort of account in action in Needham’s “polythetic categories” article. Needham’s explanatory/elucidatory object in that paper was the broad question of how categories function. Like the notion of “family resemblance,” his account of the concept of a polythetic category tells us something near the maximal level of generality about the way in which language works and provides us with a tool with which to describe—but not to explain—the uses of such categories in any given and specific circumstance. So his account requires a (monothetic) definition of “monothetic” and “polythetic” categories, but from that point further requires no such definitional practice, as any given term can be described as either one or the other, the majority of course being “polythetic” and amenable to description, not explanation. He is not attempting to explain what any particular category means, so further ethnographic nuance has no particular use.

But the effect of such an account—though eminently suited to Needham’s goal for it—is to erase the level in between that of maximal philosophical generality (an explanation of how language works) and minimal ethnographic particularity (a description of how any given polythetic term is actually deployed). The only form of explanation or definition it permits is the—by definition, as it were—universal, generic explanation and definition of something as a “polythetic category.” Like the PI, its object is language as an object we already know, rather than any particular or specific problem originating in something we do not already know. It tells us about language, rather than about games, or matriliny, or marriage, or descent.

That would be one way to treat an ethnographic problem of definition, roughly corresponding to the method of “dissolving” it, against which Popper was arguing. The punch of the resolution—its dissolution—comes in rearranging our perspective on a thing we already know (“definition” as a generic activity) such that it no longer comes to appear as a problem. Martin Holbraad’s work on Cuban divination is a highly convincing anthropological example of exactly this sort of approach to questions of definition, about which I have written elsewhere (Heywood 2018a, 2018b; Holbraad 2009, 2012). This is not the only way in which one might treat such a problem, however, as I suggest later. An alternative, and more straightforwardly explanatory, approach would look for resolution in the concrete specifics of the problem.

“Dogs and I Do Not Know What Else”

The specific problem I deal with here is a fairly general one, rather than one that is highly ethnographically specific, although it does arise from my current fieldwork. That generality does not change its form as a problem to be resolved empirically, however; it merely expands the relevance of that resolution, I hope. As I have been suggesting, one consequence of the foregoing approach to anthropological knowledge—quite explicit in Leach (1961), for example—is that it erases the possibility of what we might think of as “restricted generalizations,” or indeed “comparison” of a certain form (cf. Candea 2018, 101–103, on “caveated generalizations”). Both Leach and Needham, for instance, make universal generalizations (Needham in linguistic form and Leach in structural, mathematical form), and obviously they leave open the possibility of particular description in any given case. What they close off is what we might—albeit not altogether happily—think of as the level in between: any invocation of a classificatory analytic term that ties it to a particular sort of context (whatever the nature of the context) and distinguishes it from others. We cannot employ the category “matrilineal” to distinguish some societies from others. Of course, as Candea points out, we continue to do this “with a pinch of salt” all the time (2018, 209–211).

The problem I will address is one raised in my own field site, but it is certainly not confined to that context. It runs as follows: Why does it appear to many people in Europe and the United States that, as George Orwell put it in 1944, “of all the unanswered questions of our time, perhaps the most important is, ‘what is fascism?’ ” To be clear, then, my problem is not the definitional problem of “What is fascism?” but the explanatory problem of “Why do people continue to put so much energy and passion into looking for definitions of ‘what fascism is’?”

Trying to define fascism is a particularly fraught endeavor. Historical and political arguments over the proper meaning and definition of fascism have been taking place since it first emerged as a phenomenon in the 1920s and show no immediate sign of cooling off. A range of definitions have been proffered by eminent historians of the subject in search of a “fascist minimum” (Eatwell 1996), while at least one prominent scholar became so frustrated by the ambiguous use of the term that he famously called for it to be banned from historical discourse (Allardyce 1979). Orwell, in raising the question of “what fascism is,” was making nearly the same point in remarking that he had heard the word applied to “farmers, shopkeepers, Social Credit, corporal punishment, fox-hunting, bull-fighting, the 1922 Committee, the 1941 Committee, Kipling, Gandhi, Chiang Kai-Shek, homosexuality, Priestley’s broadcasts, Youth Hostels, astrology, women, dogs and I do not know what else” (1944). Historians and other academics have defined fascism as, among other things, a petit-bourgeois response to the development of socialism (see, for instance, Poulantzas 1974; Trotsky 1944), a psychological phenomenon resulting from a kind of mass hysteria (Reich 1933), a species of “developmental dictatorship” (Gregor 1979), a palingenetic form of ultranationalism (Griffin 1991), and a sort of religion masquerading as a political movement (Gentile 1990), to name only a few such definitions.

Recently these debates have become yet more fraught by, as it were, coming alive. They have moved from residing largely or entirely in the realm of scholarly journals and academic conferences into the world that such journals and conferences aim to investigate, from the abstract to the concrete, from analysis to object (Slate magazine, for example, printed an excerpt from Kevin Passmore’s Fascism: A Very Short Introduction as part of its academy series on fascism, suggesting readers consult the extract to determine whether they were living in a “fascist state” [Onion, Thomas, and Keating, n.d.]; the Atlantic, noting the “elusiveness” of definitions of fascism, interviewed Robert Paxton in search of a checklist of features to assess the extent to which Donald Trump is a fascist [Green 2016]). The pages of international news and commentary are filled with speculation as to whether and how far France’s National Front, Germany’s Alternative for Germany, or Austria’s Freedom Party “count” or do not “count” as “fascist,” and the word was even in the running to be Merriam-Webster’s “word of the year” in 2016.

Fascism is in some sense an obvious candidate example of a family resemblance term or a polythetic category. Since its coinage at the end of the First World War, it has provoked virulent argument and debate over what exactly it means and how it should be defined, debate that shows no signs of abating at present, and which is in fact increasing in volume as various contemporary political figures and movements are labeled with the term. Part of the reason it provokes such debate is that such figures and movements, both past and present, do not appear to share any single predicate that qualifies them for membership in the class (see, for example, Allardyce 1979). Of course, this might be said to be true of other political movements (and I have described “communism” in Bologna in a similar fashion [Heywood 2015]). But the problem is exacerbated by a number of factors in the case of fascism, including the lack of any clear doctrinal text, inconsistency of practice and policy on the part of “fascist” regimes, an apparent aversion to ideological or theoretical self-definition on the part of self-declared fascists themselves (at least some of whom might well have approved of the notion of fascism as a “family resemblance”), and the peculiar fact that fascist movements have been, if they have been anything, usually ultranationalist in character while also—arguably—forming a supranational object of some form.²

All of these factors, as well as more traditional problems of definition, combine to make it extremely difficult to define fascism, while apparently doing nothing to dispel—and indeed perhaps fueling—the appetite of historians, political scientists, journalists, commentators, and ordinary people for attempting to do so.

One such attempt is an article in the New York Review of Books by Umberto Eco (1995), which lists many of the aforementioned problems with defining fascism and concludes that it is, in fact, an excellent example of a family resemblance term. The term family resemblance also recurs frequently in modern newspaper and magazine accounts of the rise of the Far Right in Europe and the United States, in arguments both for the fact that Donald Trump and Marine Le Pen are in some sense parts of the same phenomenon and for their further inclusion within an “ur-fascism” that would also cover its historical manifestations (e.g., Esposito 2015; McDougall 2016).

Eco’s specific suggestion is that, like game in Wittgenstein’s original argument, “fascism became an all-purpose term because one can eliminate from a fascist regime one or more features, and it will still be recognizable as fascist” (1995). There is clearly a sense in which fascism is indeed an “all-purpose term,” the same sense that makes Eco’s argument compelling for many historians and contemporary political commentators worried about the (mis)application of the term.

But for precisely that reason, Eco’s analogy is slightly misleading. For many of the people he is addressing, that audience of historians, political commentators, and indeed many contemporary Italians, the word fascism is not at all like the word game, in that its proper definition and application are exactly what they are concerned with. That is from whence Eco’s argument derives its power: it sounds like it is a description of how a word is being used but is in fact a prescription for how it ought to be used. It competes with, rather than explains, other attempts at definitions of fascism, in the same way in which I have suggested the concept of “polythetic categories” was explicitly intended by Needham to compete with people’s practices of monothetic definition, and in the same way in which Suits points out of Wittgenstein that he knew already that games were indefinable, without in fact having to “look and see.” It might explain “what fascism is” in ordinary language terms, but what it certainly does is to explain away why that question matters to people in the first place. It is a form of “therapy” in the sense that Wittgenstein intended philosophy to be, in that it aims at the dissolution of the problem of defining fascism by pointing to the mistaken premise of the problem—namely, that fascism has or should have a definition.

To sum up, then, I am suggesting that it is perfectly possible to see the question, “What is fascism?” as a sort of Wittgensteinian puzzle, amenable to dissolution. But I am also suggesting that such a reading risks missing something worth explaining, not about definitions but about fascism—namely, the problem of why it is that people appear so concerned to define it in the first place. In other words, it may or may not be true that “What is fascism?” is a misposed question. But whether it is or not is entirely immaterial to the question of why people we might study keep on asking it. It would be relevant if we saw our task, as Wittgenstein did, as being the therapeutic one of correcting such misposed questions. But this would be an unusual position for an anthropologist to take, to say the least, and it is not mine here. Mine, again, is the question of why the search for definitions persists in the case of fascism.

Debates of a comparable intensity and range do not go on over the proper meaning of socialism or liberalism, for example, and users of those terms—at least in English—do not usually feel the need to consult experts in order to assess whether their usage is correct (which is not to say that such consultation would be always unhelpful). One clear difference between fascism and those terms is that there are comparably fewer people involved in such debates who would define themselves with the word, though, as I will describe, there are some. One might then think that the inability to rely on oneself as a yardstick of meaning, and the comparable dearth of self-declared exemplars of fascism to whom one might turn instead, is a factor contributing to the continuation of such arguments, and this is no doubt correct, though I think not the whole story. Another important and related difference is the fact that fascist is more often used as a term of disapprobation than comparable terms, though this fact alone is again not enough to explain a craving for definition, given that everyday terms of disapprobation are habitually used without the least interest in whether they technically apply to the people they are employed to describe (someone may be a “bastard” without us needing evidence that they were born out of wedlock).

To take another academic example, one that combines the characteristics of lacking self-declared referents as well as that of being used as a term of disapprobation: neoliberal and neoliberalism are notoriously nonspecific in meaning. Far from those characteristics leading to a clamoring for clarity or extensive debates over proper usage, in other words, scholars who use the terms seem to revel in their ambiguity and the capaciousness of their referential universe (see, e.g., Muehlebach 2012). It is usually those who, for whatever reason, do not tend to use the terms who demand some greater precision of meaning (see, e.g., Eriksen et al. 2015; Ferguson 2010; Heywood 2014).

As I have described it thus far, the problem I have been treating is why fascism appears to attract a certain sort of definitional or indexical attitude. That is, a wide range of people have, since its inception as a political movement, worried or theorized about the question, “What is fascism?” as Orwell succinctly put it, and in doing so they have often proffered specific answers to that question, such as a palingenetic form of ultra-nationalism (Griffin 1991) or resistance to transcendence (Nolte 1966), to give two quite specific instances, and they have also tended to do so in full awareness of the fact that others have proffered alternative, but often equally specific, answers. They have, in other words, treated fascism as if it were, in effect, a sort of natural kind term, or a species of rigid designation.

Originating in the work of post- and, in at least some sense, anti-Wittgensteinian philosopher Saul Kripke, a rigid designator is a word that is used with what Hilary Putnam calls an “indexical intention,” in that it is designed persistently to pick out the same sort of object, regardless of the knowledge the user has about the nature of that object (Kripke 1980; Putnam 1975). The paradigm case is the way in which we tend to use proper names: roughly speaking, if I say, “Nixon was the president of the United States,” then I use the word Nixon intending to pick out one and only one particular object in the world, and then I attribute to that object a description, in this case, “was the president of the United States.” If I likewise say, “Nixon may not have been president of the United States,” my intention is to speak about a possible world in which that description does not apply, but the whole point of the usage of the name Nixon in that context is that it still denotes the same person. I use the name, in other words, to rigidly designate the same object no matter what description is also true of the object. This is also often the case with very clearly indexical terms such as pronouns (you, he, she) and words like now, today, and yesterday, which are usually not used as, even though they can be read as, disguised descriptions.

A subspecies of rigid designators is legal terms, like theft, murder, arson, and so on (Marmor 2013, 581–587). These words, when used in legal contexts, are often intended to designate a special sort of act, and no other. Newspapers refer to “alleged” crimes until those crimes have been proved in court, and we habitually refer to “joyriding,” for instance, rather than theft when discussing the crime of taking a car for the purpose of cruising around in it rather than for the purpose of depriving the owner of it (even if we do not properly understand the legal distinction). As Paul Dresch has put it in regard to the importance of categories in legalistic thinking, “One has to have an idea of, for instance, ‘kinsmen,’ and of what they should do, as distinct from experience of my particular brother or cousin, before one can complain of them not doing it. Legalism makes such categories explicit” (2012, 12). Note also how this framing runs decidedly against the grain of Wittgenstein’s arguments about generalism and mental pictures.

Fascism has a historical existence as a legal term of art across Europe. In Italy, for example, in the closing stages of the Second World War and subsequently, the Allies and the postfascist Italian government institutionalized a range of measures designed to “de-fascistize” the Italian state (see, e.g., Domenico 1991). They issued lengthy questionnaires to state employees in an attempt to evaluate exactly how “fascist” they were; they criminalized attempts to reconstitute the fascist party in the postwar constitution by specifying some of the means by which this might be recognized. In the case of the Nuremberg Trials, the rigid designation employed was even more specific, and the court declared membership of the leadership corps of the Nazi Party, of the SS, and of the Gestapo and SD to be a criminal offense in and of itself, notwithstanding other offenses they may have committed.

The success of such attempts at rigidly designating what it means to be a fascist is a different question. Much of the modern anxiety over the term no doubt stems from the sense that these attempts were highly unsuccessful, that we ought to be able to say why Donald Trump, for example, is, in fact, a fascist, and that the right form of rigid designation would allow us to do so. This would also explain the current fascination with a search for “experts” on fascism, and the hope that a historian of, say, Vichy France could provide us with a way of pointing at figures such as Trump and stating why they are fascist. I have written elsewhere of the lack of success these attempts have had in Italy, in particular, and how that explains the persistence of a personality cult of Benito Mussolini centered on Predappio, the site of my current fieldwork (Heywood 2019, 2020).

My point for present purposes is merely that the historical status of fascism as a legalistic rigid designator is a candidate explanation for a problem (“Why are we in search of definitions of fascism?”) that does not require us to reconstrue the meaning of one of its terms (definition). People look for essential and defining features of fascism because they imagine it in the same way in which they imagine murder or theft. This explanation may of course be mistaken. But to say instead only that fascism has no definition because it is a polythetic or family resemblance term is not to explain the fact that people search for definitions but to explain it away. It is to add no new information about “fascism” the concept (or the object in the world), only about fascism, the word.

Conclusion

Reshuffling our conceptual categories is no doubt frequently a helpful exercise. But it cannot be the only one we pursue. It leaves the facts of the matter unchanged, when closer investigation of those facts may yet yield valuable results. Had Evans-Pritchard left interest in Azande rationality and witchcraft at the level of a “category mistake,” we would have none of his specific insights on the particular nature of Azande thought on causality, nor the insight that that thought resembles our own in certain surprising respects.

Most anthropology undoubtedly occupies both the Wittgensteinian and Popperian modes at different points. But in order to do that, it must explain, as well as describe. Not all questions simply dissolve in good description. That conclusion follows necessarily from the fact that anthropology’s objects are not only objects about which new empirical information is irrelevant. The Wittgensteinian position on philosophy as “pure description” only makes sense because the description in question already takes understanding for granted. The famous “look, and see” injunction, as Suits argued, is really unnecessary because part of the point is that the reader already knows, in some sense, the truth of Wittgenstein’s claims, because they are descriptions of how the reader themselves ordinarily uses language.

The objects of anthropological knowledge are often both more and less than this. They are “more” in the sense that we do not possess understanding about them in advance, such that all that is required to grasp them is the correct slotting of them into place in an already-existing conceptual schema. They are “less,” if you will, because they are usually far more specific than, say, “language.” To take Needham as an example one final time: “The outcome of analyses of this kind should not be seen merely as a local or technical rectification of European academic argument, but as pointing to a general hazard of language which presumably afflicts men in any tradition when they classify their fellows and their nature” (1975, 367). The whole structure of this sort of argument obligates it to explain—or show, more generously—something universal about language, while denying it the ability to explain anything more specific, because the universal fact shown is that explanation is impossible. Anthropologists usually “look, and see” a little more closely than this.

Beyond the specific question of the enduring value of explanation, the “dissolution versus explanation” distinction outlined here may also be a useful heuristic in thinking through the sorts of questions that any ethnographer is likely to face. Resolving a problem by declaring the question to be misposed may often be very tempting, but before doing so it is worth asking not only whether in fact an empirical resolution is imaginable but also what sort of object one wishes to treat: a great strength of the dissolution orientation is that dissolving a problem often reveals something illuminating about anthropological categories, as Needham did, for example; an equivalent strength of explaining the resolution of a problem is that it is arguably more likely to yield some substantive ethnographic and empirical insight. Unlike Popper, who makes his disapproval of questions of disciplinary method and epistemology clear despite addressing such questions at length in his Moral Sciences Club paper (1952), I have no intention of suggesting that illuminating anthropological categories is an inappropriate aim. Only that it does not serve us when we confuse it with that of actually answering an ethnographic question.

NOTES

1. Acknowledgments: I am very grateful to all the participants in the workshop on explanation from which this chapter and book result, and to Matei Candea for organizing that workshop with me. I am also grateful to Simon Blackburn, Harri Englund, and James Laidlaw for their very helpful thoughts and comments on this chapter, and to the participants of the social anthropology seminar at Oxford University for their comments and questions on an earlier version of it.

2. 1. Unless otherwise noted, all citations are from Wittgenstein 1998.

3. 2. This is with the exception of the extremely short entry in the Italian Encyclopaedia of 1932, which was at least partly devoted to rejecting the notion of “doctrine.”

REFERENCES

• Allardyce, G. 1979. “What Fascism Is Not: Thoughts on the Deflation of a Concept.” American Historical Review 84:367–388.
• Archetti, E. 1999. Masculinities: Football, Polo, and the Tango in Argentina. London: Bloomsbury.
• Candea, M. 2018. Comparison in Anthropology: The Impossible Method. Cambridge: Cambridge University Press.
• Crary, A., and R. Read. 2000. The New Wittgenstein. London: Routledge.
• Delaney, C. 1986. “The Meaning of Paternity and the Virgin Birth Debate.” Man 21:494–513.
• Domenico, R. 1991. Italian Fascists on Trial, 1943–1948. Chapel Hill: University of North Carolina Press.
• Dresch, P. 2012. “Introduction: Legalism, Anthropology and History: A View from Part of Anthropology.” In Legalism: Anthropology and History, edited by P. Dresch and H. Skoda, 1–37. Oxford: Oxford University Press.
• Eatwell, R. 1996. “On Defining the ‘Fascist Minimum’: The Centrality of Ideology.” Journal of Political Ideologies 1:303–319.
• Eco, U. 1995. “Ur-Fascism.” New York Review of Books, June 22, 1995. http://www.nybooks.com/articles/1995/06/22/ur-fascism/.
• Edmonds, D., and J. Eidinow. 2001. Wittgenstein’s Poker. London: Faber and Faber.
• Eriksen, T., J. Laidlaw, J. Mair, K. Martin, and S. Venkatesan. 2015. “The Concept of Neoliberalism Has Become an Obstacle to the Anthropological Understanding of the Twenty-First Century: Debate.” Journal of the Royal Anthropological Institute 21:911–923.
• Esposito, F. 2015. Fascism, Aviation and Mythical Modernity. London: Springer.
• Evans-Pritchard, E. 1937. Witchcraft, Oracles, and Magic among the Azande. Oxford: Clarendon.
• ______. 1940. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon.
• ______. 1973. “Some Reminiscences and Reflections on Fieldwork.” Journal of the Anthropological Society of Oxford 4:1–12.
• Ferguson, J. 2010. “The Uses of Neoliberalism.” Antipode 41:166–184.
• Gellner, E. 1960. Words and Things. London: Penguin.
• ______. 1988. “The Stakes in Anthropology.” American Scholar 57:17–30.
• Gentile, E. 1990. “Fascism as Political Religion.” Journal of Contemporary History 25:229–251.
• Gluckmann, M. 1955. “The Peace in the Feud.” Past and Present 8:1–14.
• Green, D. 2016. “The Elusive Definition of ‘Fascist.’ ” Atlantic, December 18, 2016. https://www.theatlantic.com/politics/archive/2016/12/fascism-populism-presidential-election/510668/.
• Gregor, A. 1979. Italian Fascism and Developmental Dictatorship. Princeton, NJ: Princeton University Press.
• Griffin, R. 1991. The Nature of Fascism. London: Psychology Press.
• Gullion, J. 2016. “Show, Don’t Tell.” In Writing Ethnography, 75–78. Rotterdam: Sense Publishing.
• Heywood, P. 2014. “Neoliberal Nation? Mobbing and Morality in Italy.” Journal of the Royal Anthropological Institute 20:151–153.
• ______. 2015. “Equivocal Locations: Being ‘Red’ in ‘Red Bologna.’ ” Journal of the Royal Anthropological Institute 21:855–871.
• ______. 2018a. After Difference: Queer Activism in Italy and Anthropological Theory. Oxford: Berghahn.
• ______. 2018b. “Making Difference: Queer Activism and Anthropological Theory.” Current Anthropology 59:314–331.
• ______. 2019. “Fascism, Uncensored: Legalism and Neofascist Pilgrimage in Predappio, Italy.” Terrain 72:86–103.
• ______. 2020. “Ordinary Exemplars: Cultivating the Everyday in the Birthplace of Fascism.” Comparative Studies in Society and History 64:91–121.
• Holbraad, M. 2009. “Ontography and Alterity: Defining Anthropological Truth.” Social Analysis 53:80–93.
• ______. 2012. Truth in Motion: The Recursive Anthropology of Cuban Divination. Chicago: University of Chicago Press.
• Holbraad, M., and M. Pedersen. 2017. The Ontological Turn: An Anthropological Exposition. Cambridge: Cambridge University Press.
• Ingold, T. 2013. Making: Anthropology, Archaeology, Art, and Architecture. London: Routledge.
• ______. 2014. “That’s Enough about Ethnography!” HAU: Journal of Ethnographic Theory 4:383–395.
• Kripke, S. 1980. Naming and Necessity. Cambridge, MA: Harvard University Press.
• Leach, E. 1961. “Rethinking Anthropology.” In Rethinking Anthropology, 1–28. London: Athlone.
• ______. 1966. “Virgin Birth.” Proceedings of the Royal Anthropological Institute of Great Britain and Ireland, no. 1966:39–49.
• ______. 1984. “Glimpses of the Unmentionable in the History of British Social Anthropology.” Annual Review of Anthropology 13:1–23.
• Luhrmann, T. 2012. When God Talks Back: Understanding the American Evangelical Relationship with God. New York: Alfred Knopf.
• Malinowski, B. 1932. The Sexual Life of Savages. London: Routledge.
• Marmor, A. 2013. “Meaning and Belief in Constitutional Interpretation.” Fordham Law Review 82:577–596.
• McDougall, J. 2016. “No, This Isn’t the 1930s—but Yes, This Is Fascism.” The Conversation, November 16, 2016. http://theconversation.com/no-this-isnt-the-1930s-but-yes-this-is-fascism-68867.
• Muehlebach, A. 2012. The Moral Neoliberal: Welfare and Citizenship in Italy. Chicago: University of Chicago Press.
• Needham, R. 1958. “The Formal Analysis of Prescriptive Patrilateral Cross-Cousin Marriage.” Southwestern Journal of Anthropology 14:199–219.
• ______. 1971. “Remarks on the Analysis of Kinship and Marriage.” In Rethinking Kinship and Marriage, edited by R. Needham, 1–34. London: Tavistock.
• ______. 1975. “Polythetic Classification: Convergence and Consequences.” Man 10:349–369.
• Nolte, E. 1966. Three Faces of Fascism. New York: Halt, Rinehart, and Winston.
• Obeyesekere, G. 1992. The Apotheosis of Captain Cook: European Myth-Making in the Pacific. Princeton, NJ: Princeton University Press.
• Onion, R., J. Thomas, and J. Keating. n.d. “Fascism: A Slate Academy.” Slate. Accessed January 3, 2023. http://www.slate.com/articles/slate_plus/fascism.html.
• Orwell, G. 1944. “What Is Fascism?” Tribune (London). https://www.orwell.ru/library/articles/As_I_Please/english/efasc.
• Popper, K. 1952. “The Nature of Philosophical Problems and Their Roots in Science.” British Journal for the Philosophy of Science 3:124–156.
• Poulantzas, N. 1974. Fascism and Dictatorship: The Third International and the Problem of Fascism. London: Verso.
• Putnam, H. 1975. “The Meaning of ‘Meaning.’ ” Minnesota Studies in the Philosophy of Science 7:131–193.
• Reich, W. 1933. The Mass Psychology of Fascism. London: Souvenir Press.
• Richards, A. 1957. Chisungu: A Girl’s Initiation Ceremony among the Bemba of Northern Rhodesia. New York: Grove.
• Ryle, G. 1949. The Concept of Mind. London: Hutchinson.
• Sahlins, M. 1985. Islands of History. Chicago: University of Chicago Press.
• ______. 1995. How “Natives” Think about Captain Cook, for Example. Chicago: University of Chicago Press.
• Schneider, D., and K. Gough. 1961. Matrilineal Kinship. Berkeley: University of California Press.
• Spiro, M. 1968. “Virgin Birth, Parthenogenesis, and Physiological Paternity: An Essay in Cultural Interpretation.” Man 3:242–261.
• Suits, B. 1978. The Grasshopper: Games, Life, and Utopia. Toronto: University of Toronto Press.
• Thornton, R. 1985. “ ‘Imagine Yourself Set Down …’: Mach, Frazer, Conrad, Malinowski, and the Role of Imagination in Ethnography.” Anthropology Today 1:7–14.
• Trotsky, L. 1944. “Fascism: What It Is and How to Fight It.” Marxists Internet Archive. https://www.marxists.org/archive/trotsky/works/1944/1944-fas.htm.
• Wacquant, L. 2004. Body and Soul: Notebooks of an Apprentice Boxer. Oxford: Oxford University Press.
• Weiner, A. 1976. Women of Value, Men of Renown. Austin: University of Texas Press.
• Wittgenstein, L. 1922. Tractatus Logico-philosophicus. London: Kegan Paul.
• ______. 1998. The Collected Works of Ludwig Wittgenstein. Edited by G. E. M. Anscombe, G. H. von Wright, R. Rhees, and H. Nyman. Oxford: Blackwell.