The idea of an ethically committed social science

This article presents a long overdue analysis of the idea of an ethically committed social science, which, after the demise of positivism and the deeming of moral neutrality as impossible, has come to dominate the self-understanding of many contemporary sociological approaches. Once adequately specified, however, the idea is shown to be ethically questionable in that it works against the moral commitments constitutive of academic life. The argument is conducted with resources from the work of Peter Winch, thus establishing its continuing relevance and critical importance for the social sciences, sociology in particular. Special reference is made to heretofore unappreciated aspects of Winch’s work, including within the groundbreaking The Idea of a Social Science and Its Relation to Philosophy, but focusing specifically on his later contributions to ethics.


Introduction
More than 60 years have passed since the publication of Peter Winch's classic work The Idea of a Social Science and Its Relation to Philosophy (ISS), a book that has since generated a considerable volume of debate (most famously Wilson, 1970). This notwithstanding, how deep a challenge the book really presents to the social sciences has not been fully appreciated. 1 Thinking of the book as an anti-positivist treatise, as is customary, misses that its challenge is in fact independent not only of the kind of scientific aspiration that may be at work in sociology, which is the discipline this article will focus on (that is, whether it derives from the philosophy of science of yesteryear or is newly minted), but also of the very existence of any scientistic leanings. Even sociological programmes that do not draw on renditions of scientific method provided by the philosophy of scienceor, otherwise put, do not see themselves as related to physics or biology but instead sit closely with philosophy or literatureare nevertheless equally unwilling to be guided by the book's central message: to pay close attention to the conceptual forms wherein the various kinds of understanding that make up social life find expression. A tendency towards illegitimate abstraction and the conflation of conceptual and empirical questions are, in truth, not only the prerogative of a positivist social science that is no longer; they are deeply entrenched within persistently dominant conceptions of what constitutes a sociological problem and what constitutes a finding: in sum, what it means to practice theory and what to conduct research.
Given that a number of scholars have published work in this journal (Kemp, 2003;Lukes, 2003;Lynch, 2000;Sivado, 2011) and elsewhere (Hutchinson, Read, and Sharrock, 2008) that rehearses in detail the arguments of the book, I will mostly refrain from repeating them here; I wish instead to break new ground. I will thus be concerned mainly with arguing not that ISS possesses a relevance that extends beyond the one it is usually understood to have, true though I think that to be, but that the relevance of Peter Winch's work extends beyond the topics explicitly dealt with in ISS and 'Understanding a Primitive Society' (Winch, 1964b), the two pieces by which social scientists are most likely to be acquainted with him. I will aim to show that his later work on ethics, although not directly addressing the social sciences, can nevertheless supplement his earlier arguments and have a particularly powerful bearing on recent developments in sociology.
Written in the late 1950s, ISS responded to intellectual tendencies during an era when sociology was animated by the vision of turning itself into a genuine science. While it would be an exaggeration to declare this vision now deadinforming as it does current approaches such as critical realism and analytical sociology, among othersafter the 1960s and the (official, at least) demise of positivism, a different, equally unfulfilled aspiration has mostly come to substitute itself for that of becoming a science: the aspiration of being ethically and politically consequential. Although there is nothing particularly novel about this idea, being, for example, as programmatically embedded in Marx as explicitly shunned by Weber, it was the former figure's incorporation into the sociological canon after the 1960s that paved the way for the dominance this vision enjoys as it feeds into a host of contemporary orientations in sociology.
The aspiration of being ethically and politically consequential is not only the historical successor of positivism but is also perceived by many to be its logical successor, in the sense that it is seen as entailed by the rejection of positivist assumptions, in particular the idea of a morally neutral or value-free sociology. According to a typical way of arguing, this idea refers to a mythical creature, in effect expressing an impossibility (e.g. Becker, 1967: 239;Gouldner, 1962: 212;1968: 103;1970: 485; cf. Davydova and Sharrock, 2003: 373), a fact taken to imply that the only realistic alternative to putting up 'a vain ritual of moral neutrality' (Gouldner, 1962: 212) is the pursuit of explicitly value-laden projects.
Thus, the pendulum has since swung towards projects animated by the idea of an inevitably ethically committed social science. Obvious examples of such projects are the 'emancipatory approaches', such as critical theory (see Fay, 1987;Geuss, 1981), the various forms of feminism (e.g. Collins, 2000;Fraser, 2013;Smith, 1987), poststructuralism (e.g. Butler, 1990;Laclau and Mouffe, 1985), and postcolonialism (see Bhambra, 2016;Go, 2016), but there are also other programmes that are, in part or wholly, animated by the idea and thus demonstrate its rather wide reach: 'a social science that matters' based on the Aristotelian notion of phronesis (prudence or practical wisdom; Flyvbjerg, 2001), 'a public sociology' (Burawoy, 2005) based on 'moral commitment' (Burawoy, 2014), and a sociology aiming at the production of 'transformative knowledge' (Morgan, 2016). In what follows, I will set out the idea of what I have dubbed 'an ethically committed social science' in some generality rather than discuss versions by particular authors, so it is worth stating clearly at the outset that there are potentially important differences that may require further care, which I will not be providing here. Nevertheless, I believe that the idea of an ethically committed social science is held in common in its basic form, which I will seek to reconstruct, and can be discussed as such. However, this is not to say that it is held in common by all sociological programmes. Not only are there important national differences between sociological traditionsfor instance, between the United States and Britainbut there are indeed numerous programmes that do not organise their aims around this idea, such as Bourdieusian sociology, social network analysis, demography, and cultural sociology, among others. Nevertheless, it is worth insisting that the idea maintains a significant level of disciplinary penetration, including when used as a shibboleth among members and students. 2 Although a frequently employed type of reasoning understands ethical commitment as a matter of necessity, it is worth pointing out that there are other options available. To see this, consider Martyn Hammersley, who, having written extensively on methodological matters in the social sciences, addresses the analogous question 'Is research political?' (Hammersley, 1995), understanding 'political' as referring to power and values and, thus, as close to the idea of ethical commitment. Instructive for our purposes, besides the potential overlap in topic, are the different senses Hammersley understands the question to possess. The strongest version amounts to the theoretical question of whether it even makes sense to suppose that research could ever be anything other than political, itself close to the sense of ethical commitment as necessary, as already mentioned. But he also singles out, given a negative response to the theoretical question, the empirical one of whether research is or is not in actual practiceand, furthermore, the normative one of whether it ought or ought not to bepolitical. Returning to ethical commitment, we thus end up with the following range of questions: can social science (not) be ethically committed? Is it? Ought it?
In this article, I would like to examine what these questions involve, as well as the different senses of 'ethical commitment'. I will be making use of materials from Winch to piece together and defend an original argument towards which he might have been sympathetic but which he did not articulate to this degree. As Winch notes in Ethics and Action (1972: 1), the fact that we may come to say seemingly inconsistent things about a question when we approach it from different angles is instructive. My purpose in what follows is to use this fact in order to help us find our feet with the idea.
To think through these difficult issues, then, my strategy will be to start in the following two sections from some aspects of Winch's work that, either directly or when treated by commentators in a certain way, lead us to consider the idea of an ethically committed social science as a necessary one, and that thus work together with, rather than against, various sorts of 'impossibility argument'. I will consider the claims, first, that accuracy and justice demand a commitment to a morally resonant vocabulary and, second, that morality is not a contained 'form of activity' and, therefore, social science cannot choose to remove itself from ethical commitments. I will then direct my attention to the idea of an ethically committed social science in order to offer a clarification of its structure. This will consist in sketching out three of its basic features: commitment by virtue of membership in the discipline, unanimity, and partisanship. I will, finally, proceed to show, taking into consideration further aspects of Winch's work on ethics, where the idea runs into distinctly moral problems, thus also exhibiting what Winch's response might have been to the idea of an ethically committed sociology, which succeeded the idea of a social science, about which he certainly had important things to say.

Our morally resonant vocabulary is necessary for accurate and just description
Concepts play a constitutive role in social life. This fact bears the implication that demands of accuracy bind the descriptions of the social scientist to the conceptual forms of the society they are studying (see Tsilipakos, 2020). Winch illustrates this point in various ways, including with reference to the economist's second-order notion of liquidity preference, arguing that its validity depends on its being parasitic on autochthonous terms such as money, selling, and assets, and the concepts they express. Being committed to a vocabulary that has a constitutive role comes with yet further baggage, especially if that vocabulary is implicated in the moral order that is social life. The epigraph to ISS stands as a clear expression of the fact that there are demands of justice riding on those of accuracy (cf. Gunnell, 2009). Winch comes back to this in his later work. In 'Particularity and Morals', he considers the respect due to a person as something tied to 'taking seriously their description of their action, and this is also treating them justly or with justice, in treating seriously their understanding of the situation and of what the situation calls for them' (Winch, 1987: 177). There are a number of senses of injustice here, including that of discarding personal moral dispositions, to which I will return when discussing the issue of uniform commitment across the discipline. For the moment, let us look into the injustice of not taking seriously a person's description of their action.
In his introduction to a subsequent edition of ISS, Raimond Gaita (2008), folding in his knowledge of Winch's later work as well as his own concerns on ethics, examines the issue of description with reference to the preservation of the ethical resonances afforded by a constitutive vocabulary. He invokes an imaginary example of a social scientist who, in studying, as he puts it, 'discipline procedures' for improving student conduct in a school, wants to base her description of what the teachers in the school are after on a neutral notion of 'behaviour modification', while dispensing with, or perhaps merely overlooking, the fact that some teachers might describe these proposals as an attempt at bribery, or might object to punishment that relies on the involvement of the rest of the student body and may thus result in the humiliation of those punished. Gaita's point is that opting for the notion of behaviour modification does not allow an understanding of what bribery, punishment, or humiliation mean. He takes the verb to mean to be crucial in the sense he exemplifies via the following construction: 'Don't you understand what it means to be humiliated in the way children would be if you encouraged their fellow students to disapprove of them?' (ibid.: xxv). In speaking thus, Gaita is already invoking Winch's own remarks on the sometimes internal relation between being willing to consider (or being indifferent to) an action and really understanding what that action means. 3 Yet the sense of 'understanding what it means' in Gaita's formulation is not entirely lucid, particularly regarding the extent to which it has to do with understanding a form of words.
First, let us try to clarify the matter by thinking of some appropriate replies. For example, someone could respond with the following words: 'Oh, yes, I remember being humiliated as a child. If anyone knows, I ought to know; I have had that experience and I know how unpleasant and demeaning it is', or 'Yes, I do understand what it means and how bad it can be, which is why I see that it is imperative that we do not allow it to happen.' But consider the following alternative response: 'Well, I know what it means; I have had the experience and I have seen how useful it is in life, which is precisely why I think it will make tough kids out of them.' All of the above are possible moral understandings expressed towards what is happening. 4 They exhibit the intended sense of the expression 'what it means' in connection with behaving in certain ways, and they do so in large part through the invocation of one's personal experience and how it has shaped their moral understanding of the situation.
It can now be appreciated that there are important differences between the question of 'what the words "punishment," "bribery" or "humiliation" mean' and the question of 'what it means to punish, bribe or be humiliated'. When asked in the abstract, the latter somewhat converges with the former in so far as it relates to seeing that these terms are used to describe behaviour to which disvalue can attach. But the questions diverge in sense, and moral differences emerge when the issue of understanding what bribery, humiliation, or punishment mean is posed not in general but in connection with particular cases and one's relation to them as a person. As we saw, the operative sense of 'understanding what it means' is that of sharing a set of experiences and a moral outlook rather than seeing that, for example, in general, bribery is something to be disapproved of, or being competent in the use of the term bribery. This competence, of course, is something presupposed in any further moral discussion.
Thus, to return to the crucial aspect of Gaita's example, since the teachers are faced with assessing the ethical implications of certain courses of action, the attempt to capture the issue with the term behaviour modification, apart from importing particularly manipulative connotations, shuts down as opposed to making available or deepening in any sense understanding of the problem, the range of possible responses, and, in particular, what they mean. 5 There are moral features inextricably bound with the situation and, for Gaita, sticking with the terms bribery, punishment, and humiliation, and paying attention to the tone with which they are uttered, is necessary if those features are to even register for the social scientist. Gaita (2008: xxvi-xxvii) proceeds to argue that we need to understand what is deep or shallow in the teachers' response by employing evaluative or 'thick concepts', 6 which require distinguishing between real and counterfeit forms of, say, justice by appeal to art and to 'extraordinary language'. 7 This, however, seems to run together the idea of registering the different distinctions between what is real and what is counterfeit and the idea of being able to settle between conflicting versions of them. In other words, Gaita's words might reasonably be taken as confusing the social scientist's description of the moral options and the quite separate matter of whether she opts, in this case, for the view that sees humiliation as an insuperable obstacle or as a minor inconvenience on the road to building tougher kids. 8 Indeed, both of these views are possible, each distributing depth and superficiality differently, and although commitment to a morally resonant vocabulary is necessary for both, it necessitates neither.
Returning to the overarching question of the ethical commitment of social science, Winch and, in part, Gaita are arguing in favour of the necessary commitment to the vocabulary that is constitutive of social life, if social science is to be accurate and just in its descriptions and not lose sight of moral visions that play an integral part in social situations. 9 We might describe this as a kind of 'impossibility argument', one ruling out that it might make sense to discard our morally resonant vocabulary in favour of some other set of terms, perhaps some kind of neutral 'observation language'. Yet we have seen that no part of this argument implies the necessary commitment of the social scientist to a specific moral vision.

Morality as a general 'form of activity'
Social life is a moral order, traditionally conceived so in sociology, which not only studies but is itself part of social life. This relationship raises a second set of 'impossibility considerations' regarding the idea of an ethically committed social science: if there is no way to contain morality, then there is no way for the social sciences to be insulated from it and, therefore, they must commit on at least some of its demands.
Let us approach the issue by looking at Winch's preface to the second edition of ISS (Winch, 1990b), where he casts a retrospective eye on the work, offering clarifications and qualifications and elaborating on some of the things he could have done differently. In ISS, Winch had talked about logic as embedded within activities, and about criteria of intelligibility as something pertaining to each activity, while arguing against the intelligibility of judgements as to the rationality of, first, entire activities and, second, particular stretches of conduct within an activity (for example, praying to God within religious life) on the basis of criteria imported from another area of conduct (such as scientific experimentation). Winch considered his not paying more attention to the logic-idealising flavour of formulations of this kind as responsible for a particularly unwelcome implication some read into his discussion, namely, that activities are separate and completely isolated from each other. 10 In the preface to the second edition of the book, he is, first of all, eager to complicate the understanding of 'forms of activity' or 'modes of social life' as not all occupying the same footing. In other words, he sees these expressions as applying to a more heterogeneous set of categories beyond the typical set of, for instance, religion, science, and art. Moreover, he stresses the possibility of not only overlap but, even further, internal relations between them in the sense that in some cases, 'one cannot even be intelligibly conceived as existing in isolation from others' (ibid.: xv). The two issues of the relationship between different activities, on the one hand, and of the personal commitments of individuals whose life encompasses these activities, on the other, are raised throughout Winch's work. 11 In the preface, however, he specifically points towards the understanding of morality in his essay 'Nature and Convention', published a year after ISS. This essay will be useful, then, in discussing the sense in which the notion of a 'form of activity' can apply to morality, as well as what it might mean to speak of internal relations between morality and other aspects of life.
In looking into the matter, we might first bring to our attention, so as to render more explicit, an implicit understanding that is already present in ISS. In Chapter 2, Winch (2008: 43) connects the problems of calling into question the entire idea of 'a reason for an action' to the discussion in Chapter 4, where he argues against Pareto's tendency to describe entire 'forms of activity' or 'modes of life', such as religion, as illogical. The fact that the notion of a reason for an action is described as a form of activity is particularly interesting because it shows that reason-giving and religious activity are both vulnerable to the same philosophical misconception, independently of any implication relating to their insularity. Indeed, the idea of a reason for an action and the attending practice of reasongiving are immune against being understood in any self-contained way (any more than asking for something, apologising, or promising can be compartmentalised). Moreover, giving reasons possesses a generality of such a kind that makes it perfectly intelligible to assert internal relations, for instance, with the practice of science, politics, or the law. This broadening of 'form of activity' provides a platform from which to understand morality, not analogously to science but analogously to reason-giving.
Turning now to 'Nature and Convention', Winch is there concerned to criticise Popper's dualism between decisions and facts and to point out that statements of fact and decisions, and the relation between them, are characterised by pluralism in 'science, morality, business, law, politics' (Winch, 1972: 57), seemingly confirming the unfortunate assimilation of morality to this standard categorial group. But he then comes to complicate that understanding in interesting ways. First, he notes that the notion of a scientific community differs from that of a moral community in that 'society would not be possible without [it]' (ibid.: 58). In other words, one could say that scientific community implies but is not implied by moral community, which is shown to be internally related as a presupposition to scientific activities. 12 Winch further notes: Morality cannot be called, in the same sense as can science, a 'form of activity'; it is not something one can choose to engage in or not at will. It would hardly make sense, for instance, for someone to say he had spent six weeks working hard at morality (unless this meant something like moral philosophy). (ibid.; original emphasis) The notion of working hard that might apply to things like giving promises or telling the truth, one might add, is not in the sense of working at one's skills or technique; rather, what is typically meant is working on or even struggling with oneself in order to keep a promise or to tell the truth. Winch goes on to elaborate that it does not make sense to suppose that one may choose whether to be involved with moral concerns or not. Such concerns force themselves upon us and, even when they may apparently not, a lack of sensitivity to them might leave us exposed to criticism. As Winch observes, driving home the logical distinction between 'form of activity' as applied to morality and to science: 'You cannot put yourself outside the sphere of moral discourse by saying it does not interest you. But a man [sic] who refuses to concern himself with scientific issues does not thereby expose himself to scientific judgement' (Winch, 1972: 59; original emphasis), the point being that unlike scientific judgement and its relatively contained relevancegiven that it has to do with technical matters, which allows it to be left to the expertsmoral judgement is not a self-contained or local aspect of our lives that can be avoided by someone professing lack of concern.
Having paid attention to the subtleties around Winch's understanding of the notion of 'form of activity', we can appreciate some of the relevant features of morality. We saw that it makes no more sense to speak of morality as a separate sphere than it makes sense to speak of making promises, deliberating, or thinking about one's conduct as constituting a separate sphere. This is not to say that Winch is advocating a conception of morality as all-encompassing, which, again, does not seem to make any more sense than saying that promising or deliberating are all-encompassing. Undoubtedly there is generality present, but the generality of moral concerns exhibits itself in the fact that they are intertwined with our various activities, not in somehow containing everything that goes on in those activities. 13 Moral demands may crop up in many different contexts not because these contexts are inside the sphere of moralitynor are they outside itbut because, one might say, morality is an omnipresent dimension of social life. I have stressed the possible analogy between morality and reason-giving, which lies not in any idea that morality provides us with reasons for acting (Winch, 1972: 177) but in morality's being a possible dimension of appraisal and having to do with the application of standards to what we do.
In some of his early work, Winch articulates a conception of morality as arising from our common life and, as already noted, also presupposed by it. In 'Nature and Convention', he argues that 'the social conditions of language and rationality must also carry with them certain fundamental moral conceptions' (Winch, 1972: 60-1; original emphasis), an example being that of truthfulness as a moral virtue, which, he thought at the time, stands as 'a necessary background condition in any society in which it is possible for anyone to make true statements' (ibid.: 63). He develops this line of thought further in 'Understanding a Primitive Society' (Winch, 1964b), where he observes that not everything passes as a 'moral' concern, pointing out that our human life is given shape by 'limiting notions', such as birth, death, and sexual relations, and that their various accompanying conceptions are closely linked to concepts such as good and evil. One should not conclude from this that for Winch morality derives from 'nature'. In 'Human Nature', published in 1969, he rejects the idea that morality is a way of fulfilling 'human needs', which would allegedly stand as its criteria, arguing instead that moral dispute is already involved in finding out what those needs are. Nor does Winch view favourably the subordination of morality to rationality or Kantian universal law. Moreover, Winch points out that the fact that people may disagree about what is and what is not a moral issue leads to reflection not only on our shared life together but also, importantly, on how a particular person may 'arrive at a moral understanding of [their] own life' (Winch, 1972: 3-4)a point that we will have reason to emphasise in connection with the issue of disciplinary unanimity.
All in all, it would be ill-advised to press towards a general definition of morality or of what makes a consideration into a 'moral consideration'. Winch's relevant observations, for example, that moral disgust is such by virtue of its relation to concepts such as 'betrayal, breach of faith, etc.' (Winch, 1987: 172), should be understood in the context of his explicit statement that there is a 'very wide variety of situations, arguments and judgements to which it is natural to apply the term "moral"' (Winch, 1972: 159) and, moreover, that moral judgements are themselves heterogeneous (Winch, 1987: 185). I would add that moral considerations are best understood by paying attention to their occasioned character, including their relation to a set of concepts that includes good, evil, bad, right and wrong, harm, virtue, integrity, and very many others. It is important to remember that our final court of appeal is not any definition or theory of morality, but the fact that we may or may not be capable of registering and responding to a moral outlook (cf. Winch, 1972: 200) or issue, as it may arise diffusely in our shared life, for example, in connection with the punishment of students.
To return, then, to the overarching argument, Winch's thinking can, once again, be treated as congruent with negative responses to the question of whether it even makes sense to suppose that social science could be otherwise than ethically committed. Based on what we have seen, it is not possible to distil a social science that is removed from the moral fabric of social life.
Yet, once again, caution is required. There is considerable distance between taking the failure of purging operations to imply, on the one hand, that moral demands will crop up and, on the other, that a specific set of ethical commitments or values are to be recommended. While strong enough as demonstrations against a neutral 'observation language' or a practice completely insulated from morality, impossibility arguments have usually been interpreted as directly and inescapably leading to the specific forms of ethical and political commitment that various authors wish to insist on for the social sciences. The connection is natural but, I submit, too quick, and I think Winch would agree. For example, the mere fact that social scientists share in a language that is morally resonant does not by itself remove the meaningfulness of distinctions between being dispassionate or thoughtful and partisan or fanatical. Moreover, although it may not be possible to say that a certain activity will be insulated from a particular demand, not all moral demands are constitutive of that activity. The social sciences as an academic pursuit are constitutively intertwined not with every conceivable but with a specific set of basic moral commitments. As we will see, the demands of, for example, intellectual honesty, sobriety in assessment, and dispassionate fairness, far from being part of an underhanded attempt to evade moral commitment, are indeed moral demands, and, in relation to academic life, constitutive ones at that.
In light of these remarks, I will now turn to explicitly consider in what sense the idea of an ethically committed social science, in its root form, informs and is intended by contemporary projects. It will be seen that it not only extends far beyond the cautionary remit of impossibility arguments but also ends up undermining many of the constitutive moral demands of academic life.
What does the exhortation to ethical commitment mean?
Typical proposals for an ethically and politically committed social science are offered as inescapably flowing from versions of impossibility arguments and, furthermore, as easily eliciting assent because they are thought to express an obviously commendable position that no one, when given the choice, would wish to contradict. 14 Part of the commendation derives from the absurdity of a person who would choose to be unjust, unreasonable, or irresponsible as opposed to aiming towards their own and others' enlightenment and emancipation (Geuss, 1981: 2). Another part, again hardly requiring justification, derives from the idea that in the face of pressing issues in the contemporary world, being committed is surely superior to being disinterested: why, for instance, would anyone choose to remain uncommitted when the mere fact of not taking a stance on the realities of commodification (Burawoy, 2014: 283), climate change, war, social inequality, or injustice (Collins, 2015), to name just a few issues, is obviously a moral failing?
Despite the implied self-evidence of all this, the matter is not straightforward at all. While there is no reason to deny that being ethically unresponsive can be a moral failing, it is worth wondering what kind of moral failing can logically give rise to the exhortation to be ethically committed, and, especially, in what sense social scientists can follow this exhortation.
Starting from what are untenable options, it seems difficult to conceive the exhortation as a response to an overall lack of ethical commitment that social scientists considered as individuals may have exhibited, at least not in comparison to anyone else. It is most reasonable to suppose that, as individuals, they are no different to any other group of people who may take a stance in relation to ethical matters as these crop up in their life. It would, after all, be rather implausible to claim that social scientists are particularly prone to being uncommitted in this sense on account of their character or the people they happen to be. Nor can the force of the claim be construed as concerning the idea that, apart from responses to their private issues, those individuals have no clear stance on public issues where there are ethical stakes, and ought to develop one. Once again, this would render them no different to anyone else; it would merely confirm their membership in civil society. 15 And while it is conceivable that the call could be addressed to raising the overall level of ethical responsiveness within civil society, this tells us nothing about sociology's role in this.
Instead, the exhortation really has to do with the idea that social science, sociology in our case, should be designed to play a special part in ethicalin particular public and politicalissues (Burawoy, 2005). It should not be silent about them, in the sense, for example, in which solid state physics is silent about the injustices of the penal system, but rather consider them as issues on which it ought to be highly relevant and, indeed, vocal. Naturally, I do not mean to imply that the commitment of sociology is separate from the commitment of sociologists. Those who are to assume and, moreover, express its position are in fact sociologists. Yet they are not to do so as individuals, that is, on account of certain independent aspects of their lives, but rather as members of a discipline who derive their commitment precisely from their disciplinary participation and whatever that is thought to be capable of affording (Burawoy, 2014). 16 Commitment through membership leads to a further aspect of the idea of an ethically committed social science, one that is perhaps left unarticulated: there should be unanimity among members. In other words, the point of the exhortation cannot be to encourage varied participation in public affairs, where what is thought to matter is that one participates in seriousness, much like it matters that one votes responsibly, as it were. The idea is not that social scientists will just exercise their voting rights, without it being of concern whether some vote, say, for the left and others for the right; or, to return to Gaita's example, that all teachers will become involved in the issue regardless of whether some will say that the possibility of a child experiencing humiliation is to be avoided while others that it will prove a useful character-building experience. Instead, there is the implication that social scientists will vote the same way by virtue of the fact that they derive their commitment from the discipline. Furthermore, it is presupposed that, at least most of the time, which position is to be favoured is not left open to be decided after deliberation; rather, the issue is settled in advance. 17 For instance, the vote is to be given to the left; the conundrum the teachers are faced with is best seen as a case of children being exposed to humiliation, which thus ought to be prevented.
Based on the above, then, the idea of an ethically committed social science may be unpacked as most fundamentally involving social scientists (a) being committed in their capacity as members of a discipline; (b) standing unanimously by virtue of that membership; and (c) being self-evidently in support of a position with a specific ethical and political identity. These three features, as already hinted at, work within the understanding that ethical commitment is accompanied by a measure of effectiveness (see Geuss, 1981). The idea is that it is important to heed ethical demands not merely for their own sake, as far as one's conscience is concerned or before one's peersby Winch's lights, not a bad audience at all (cf. Phillips, 1997: 289;Winch, 1996: 20) but because to do so is to bring about change, to make a practical difference. 18 A possible Winchian reply to the successor idea Now that the conception of an ethically committed social science has been set out somewhat generally, what I hope to achieve in the remainder of this article is to consider in more detail each of its features. I will invoke Winch's later work on ethics in order to deepen our understanding of their moral significance and expose some potential problems.

Commitment by membership?
What is it about disciplinary membership that is thought to result in (unanimous) ethical commitment? The question is anything but straightforward given disagreements about how to conceive of the social sciences, which may, in turn, lead to further disagreements on what exactly membership is thought to afford. It pays to consider the several options available: A first response trades on thinking of social science disciplines as accumulated bodies of knowledge; in other words, as sets of findings that have been established to the satisfaction of a community and are conceived as including sets of ethical commitments. Yet, apart from unconvincingly placing such a body of accepted knowledge beyond the fundamental disagreements that run deep in sociology, this picture seems to imply that ethical commitments are a matter of empirical knowledge, as though it were a case of sociology being capable of and concerned with demonstrating that a value is worth committing to.
A second, no less natural, response is to think of what is afforded as various special means of finding things out. Under this conception, a discipline provides its members with tools that furnish them with new powers and abilities or augment existing ones. This conception is most obviously appropriate to technical disciplines, but, depending on how open we are on what is to count as an extension or augmentation of ability, it can be applied to the social sciences too. Yet it is difficult to see what kind of ethical tools the social sciences are supposed to afford. More importantly, this view seems to take ethical commitment as a technical matter to be dealt with by a set of experts, when in fact, as argued, it is an omnipresent feature of human life whose performance could not hinge on the existence of specialist tools.
A third response identifies disciplinary affordances as the particular skills that one is trained in. These may be thought of as those abilities necessary in the use of tools or perhaps as resembling techniques or methods themselves. Under this conception, moral skills are construed as specifically targeted for development or enhancement. But such a view is more closely suited to training under a regime of, say, religious asceticism rather than under a university curriculum. Moreover, it remains difficult to see ethical commitment as something possibly resulting from these skills, as such a view does not depart from the identified erroneous conception of morality as something one is technically trained in.
In contrast to the above three conceptions, it seems to me much more appropriate to talk of disciplinary membership via the inclinations, sensibilities, and commitments that are constitutive of a discipline through being presupposed but also nourished by its culture and its aims. In doing so, it is important to acknowledge that, even though it may be difficult to support the thesis that ethical commitment stems from specialist knowledge, tools, or skills, an academic community can still be the bearer and cultivator of certain commitments that enable the exploration of possibilities that may not have been open to one beforehand.
In 'Human Nature', Winch discusses R. G. Collingwood's autobiographical account of his encounter at a very young age with Kant's moral philosophy. Collingwood describes this experience as a revelation, upon which he discovered his calling. Winch points out that Collingwood did not only discover philosophy but found out something about himself that day. Putting the point in this way is important because Winch wants to emphasise the relation between the type of responsive attitude that Collingwood exhibited and the object of that attitude. 19 The relationship, he wants to say, is internal in that the responsive attitude is specifiable only through the object, the particular subject matter and problems Collingwood found himself responding to. In other words, Collingwood found out that his inclinations were or could be developed into philosophical ones, that he was the kind of person for whom philosophy's problems and constitutive commitments had resonance, and he chose to pursue them further through participation in the discipline. 20 'Just look at how important these questions are', he might have said, much like someone being struck by the miraculous in nature might say, looking at a blossom, 'Just look at it opening out' (Wittgenstein, cited in Winch, 1987: 148).
In a previous section, I referred to the constitutive ethical commitments of academic activities. To do justice to their variety in connection with social science disciplines, one might point to, apart from general academic commitments, those that are part of disciplinary aims and assumptions and, even further, to the overall culture of a discipline as reflected in things like its concerns and worries. In sociology in the West, we might identify in this category a tendency to side with the underdog, or to worry about inequality and the oppressed, which is why most sociologists would vote for the left and, perhaps, answer Gaita's student punishment example in the way I have suggested.
There is no denying that these concerns are in many ways presupposed in the culture of the discipline and nourished by many of its activities. Yet arguments for an ethically committed social science that prioritise these concerns fail to appreciate the extent to which such commitments may work against or, in the worst case, eclipse the no less ethical commitments that are distinctively constitutive of academic life.
Winch discusses precisely this matter as part of his introduction and brief review of contributions to the highly relevant The Political Responsibility of Intellectuals (Maclean, Montefiore, and Winch, 1990). In identifying themes running throughout the volume, he notes, most centrally, the fact that partaking in and having 'chosen a certain [intellectual] life' commits one to particular values, such as truth, honesty, and integrity, which present a moral demand to be served and defended when under attack (ibid. : 13) and which, furthermore, depend for their realisation on certain cultural and institutional arrangements. Moreover, those same cultural and institutional arrangements demand to be defended when threatened, including by 'socio-political forces' (ibid.: 2). Importantly, Winch also notes that the constitutive values in question are not compatible with every other value under the sun and may, therefore, come into conflict with 'a wider political, or quasi-political, responsibility' (ibid.: 6).
Returning to the question of commitment by the fact of membership, we might thus conclude that there is indeed such a thing, not because it is dictated by a set of findings, but because membership itself depends, among others, on a commitment to a set of values constitutive of disciplinary and, more generally, academic life. Typical arguments for an ethically committed social science fail to perceive the full importance of this fact and bypass the real issue, which is not whether it is possible or not to be uncommitted but exactly which constitutive moral demands one acknowledges as a member and how one is to respond in the face of potential conflicts arising between demands.

The question of unanimity and its ethical limits
Apart from being referenced in connection with the discovery that his sensibilities were distinctly philosophical, Collingwood makes a later appearance (Winch, 1997) as a full-grown scholar who reports feeling alienated from some of his colleagues, something Winch and, I take it, most academics can relate to. It is indeed important to remember that this is a possibility even within closely knit fields. Otherwise, the fact that constitutive values are internally related with personal sensibilities might be taken as necessitating the idea that members of a discipline are to assume uniform positions. But a commitment to certain commonly held values does not preclude that further personal sensibilities will enter into how those values are seen to matter. Most importantly, a commitment to values is not identical with, nor can it automatically result in, the making of a decision or the taking of a position in a particular situation. In fact, there are ethical limits as regards the kind of unanimity that can be embraced, especially concerning specific issues and cases where there is conflict of values.
Winch preoccupied himself with the problem of what it might mean to find out for oneself what is the right thing to do. In 'The Universalizability of Moral Judgements', he positions this finding-out between decisions, on the one hand, and propositions or descriptions of facts, on the other. 21 Even in such descriptions, he argues, the ways in which we set out what the situation we find ourselves in is like include much fuller reference to the people we are and our personal tendencies and inclinations: 'If we want to express, in a given situation, how it strikes the agent, we cannot dispense with his inclination to come to a particular moral decision.… Such dispositions as this have to be taken into account in applying the notion of "exactly the same circumstances"' (Winch, 1972: 169;original emphasis). In referring to this responsive attitude of the agent, Winch does so in a way transcending any commonly held disciplinary identity and whatever level of commonality there may be on account of a common culture of sensibility. 22 An individual's personality and biography and the way they have resolved to lead their life all enter into how they are struck by a situation and what it means for them that something is the right thing to do. The idea, then, that this process can be somehow overridden or presumed to be automatically carried out through a uniform commitment to particular values or principles is not only mistaken as an account of ethical life, but is also morally questionable because it leads to superficiality, among other vices.
Underlying this idea is arguably another, namely, the conception of morality as a set of principles whose function is to guide conduct towards some kind of summum bonum, such as the free, equal or just society. Winch challenged this idea especially when seen to refer to a process that occurs in an unequivocal or uniform way. In 'Moral Integrity', he puts the point bluntly when he exclaims, 'Morality does not help overcome obstacles, if anything it puts those obstacles in front of us' (Winch, 1972: 172), meaning that if there were no moral demands, then limits to one's actions would be absent. The most important upshot of this observation, however, is that the issue of how the obstacles created are to be navigated, especially in cases where there are conflicting principles involved, e.g. liberty and equality, does not possess an answer that can be worked out by reference to those same principles but may refer to further values and, ultimately, to the person, the way they are struck by the situation and the decision they can resolve to support. 23 In cases of conflicting values and moral dilemmas, there may be no room for the idea that there is a unanimously right decision to be taken (cf. Winch, 1987: 186-7) but rather several obstacles to be handled according to conscience.

The question of partisanship
We have seen that it is not possible to sustain the presumption that working out the right thing to do can be delegated to disciplinary resources, be they a corpus of knowledge, a set of tools or skills, or even the commonly held constitutive commitments. 24 Pursuit of truth, intellectual honesty, and integrity govern academic life in an open-ended way that leaves room for individuals to work out where they stand on a particular occasion or topic. Open-ended values work together and indeed may require on behalf of individuals, as Winch points out throughout his work, an attitude of seriousness and reflectiveness when encountering complex situations where these values need to be coordinated with others and decisions made about their force and range of application.
Things are radically different when social science is governed by commitment not to open-ended values but to particular positions: open-ended values themselves as well as the process of moral seriousness are both jeopardised. A case in point is an avowedly partisan social science that defines itself as morally virtuous against a set of enemies, abstract and concrete: neoliberals (see Ferguson, 2010), anti-abortionists, antiimmigrationists, conservatives, and even the 45th president of the United States (Fasenfest, 2021) are candidates here. However opposed one may be to what these positions or individuals stand for, basing social inquiry on commitment to a contrastive position that is held to occupy the moral high ground (see Tsilipakos, 2017) and to be self-evident is ethically questionable.
In the context of controversy, Winch sees adopting a difficulty-free position as an illusion. He argues that in serious engagement with complex situations, there is a significant part to be accorded to the acknowledgement and consideration of the problems that one's own position may involve. In 'Particularity in Morals', he notes: One way in which a man may exhibit reason in the context of moral disputes is through understanding the moral positions of others opposed to his own, seeing the difficulties in them to be sure, but equally allowing them to highlight difficulties in his own position. There is absolutely no ground a priori for expecting that it will be possible to arrive at some position free of difficulties which everyone will be able to accept. (Winch, 1987: 178) We might pause to observe that the fact that Winch employs the term man generically bestows a different set of moral resonances to his statement when read in the present. It is worth pointing out that this is a facet of the complexity of the issue of a morally resonant vocabulary, which has been raised previously. Yet the point Winch is trying to make about human beings, we might say, stands. Apart from blinding one to the deficiencies of one's own position and leading one away from moral seriousness, partisanship works against the open-ended values that are constitutive of academic life. For one, Winch (2008: 95-6) speaks in ISS of philosophyand the point could readily be extended to the social sciencesas constitutively 'uncommitted inquiry' in the sense that it is not to take sides in debatessuch as that between science and religionso as to award 'the key to reality' to either, but is rather to concern itself with elucidating as best it can the different ways, including its own, of making reality intelligible. Moreover, we have already considered the demand of doing justice to how someone describes their actions and situation, which is particularly important for the social sciences. 25 Bearing in mind the significant range of difference within social life, partisanship stands firmly in the way of responding to this demand.
To reiterate, then, being uncommitted in one's inquiry and doing justice to people's description and understanding of their actions are not ways of pretending to be removed from moral demands but are themselves such demands, ones that work together with the values of intellectual honesty and the requisite reflectiveness and seriousness.

Conclusion: Winch and the history of sociological debates
Since we have travelled over a large expanse of ideas, it might be worth restating the entire argument as clearly and concisely as I can. To avoid possible confusion, I should restate that it is an argument I have constructed, articulating a position I wish to defend. It goes like this: After breaking with positivist assumptions which per impossibile held facts and values to be separate, sociology has come to embrace, to a significant extent, a successor idea, understanding itself as necessarily ethically committed. There is indeed truth in this view, but it all depends on what one understands by 'ethically committed' and, in particular, what kind of commitments are at stake. I have mustered materials from Winch in order to show how he might have handled this idea. With Winch, then, we can appreciate that there is indeed a generality to moral concerns, intertwined as they are with various activities, including social science, and that commitment to a morally resonant vocabulary that is used in social life is necessary for its accurate and just description. These insights, however, do not imply a further commitment to any specific moral vision, among the many that can be found within social life, nor do they bear the implication that social science's non-insulation from moral concerns constitutively binds it to just any of them. The idea of an ethically committed social science, as it is typically understood, moves abruptly from the fact of non-insulation and implicitly, by way of disciplinary membership, to a unanimous commitment to specific ethico-political positions. Thinking with Winch, we can appreciate not only that uniform commitment to such positions is something different in kind to being governed by the open-ended constitutive demands of intellectual life, but also that it tends to work against these demands and, moreover, to jeopardise the moral seriousness required in dealing with ethical matters.
While there may be no exhaustive list of values that are constitutive of academic life, I have maintained that truth, intellectual honesty, and integrity are central. It need not be argued that they are absolutely supreme or that they may never be judged secondary to others. The fact remains that if they were to be systematically demoted or given up, then academic life would not survive.
By way of conclusion, I would like to reinforce what has been argued about the centrality of these moral concerns by explicitly indicating how they have been both recognised and debated in a few classical and contemporary instances of sociological debate Weber, in Owen and Strong, 2004).
It has recently been reiterated by Burawoy (e.g. 2014) that it is a constitutive moral commitment for sociologists to speak against injustice, a prominent exemplary figure he cites being Pierre Bourdieu. Reflecting on this proposal, it seems to jar with the fact that some of the better-known public intellectuals who are as vocal against injustice as one would like may be linguists, philosophers, or literary scholars who do not appear or claim to derive their commitments from their disciplines. And it has indeed been argued by no less a public figure than Sartre that an intellectual 'qualifies as such by stepping outside the sphere of his or her professional activity' (Sturrock, 1998: 6). Thus, no part of the argument put forth here implies that it is not possible that sociologists might intervene in public affairs or even become public intellectuals. Of course, the issue is not that at all; rather, it is about how sociology is to think of itself, which in the past 50 years or so it has done in connection with public issue commitments. If taking a position on public issues is a moral demand on sociologists, then it needs to be evaluated side by side with the equally moral demand to consider seriously whether sociology is really a basis from which to speak against injustice (Tittle, 2004), whether there is such a thing as expertise in these matters (cf. Winch, 1990a), and whether sociologists can lead by example or 'have any right to claim to be "leaders" of any kind in matters of conduct' (Weber, in Owen and Strong, 2004: 25).
Although advocating the inescapability of an ethically committed social science, in the sense in which that idea is typically understood,  was highly sensitive to the difficulties with the idea and the attending moral demands, which render it anything but inescapable. He perceived the problems with advocating for particular factions and their positions and recommended instead a pursuit of values, a point I have drawn on in this article. He also appreciated the constant struggle required in going beyond the ritualistic assertion of commitments and acknowledging conflicts (ibid.: 133), as well as the demands placed on how one's opponents are to be handled, citing, for instance, the depiction of Persians in ancient Greek tragedies. A more recent example, we might add, is Hochschild's (2016) upright handling of Trump supporters in Louisiana. It is entirely possible, then, to avoid treating ethico-political commitments as indubitable truths and subjugating academic values to them. It is also entirely possible to strive to acknowledge 'inconvenient facts'something Weber described as an 'ethical achievement' (Owen and Strong, 2004: 20) to endeavour to obtain a sound picture of what academic life provides and what it can legitimately claim to do, while perhaps also pursuing ethico-political commitments through channels that are appropriate to them.
Given the emphasis placed on impossibilities, it may be worth ending by reaffirming that it is indeed impossible to be morally neutral and a scholar. Truth, intellectual honesty, and integrity are moral values that demand to be defended against an encroaching commodification and bureaucratisation of academic life. They equally demand to be defended against the potentially corrosive effect of political commitments that, somewhat ironically, are sometimes very well integrated with the bureaucratic institutional frameworks in place for the funding and evaluation of research.

Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author received no financial support for the research, authorship, and/or publication of this article.  (Winch, 1987: 32). 4. I am here using the terms ethics and morality interchangeably, especially in their adjectival and adverbial forms. Of course, one could and should distinguish between the two terms in various contexts, which would not, however, make a difference to the present argument. For one, Hegel famously distinguished in various ways between 'the ethical life' (Sittlichkeit) and 'morality' (Moralität). For another, ethics names a field of philosophy that may be said to have morality as its subject matter. Moreover, ethics may exclusively designate a set of rules of conduct a profession adheres to. On the other hand, it is a person's morals that we may praise or at which we may wonder. 5. Iris Murdoch lies at the background of Gaita's manner of speaking of 'knowing what a word means' and 'deepening' moral understanding. In 'The Idea of Perfection', originally published in 1964, she writes: 'There are two senses of "knowing what a word means", one connected with ordinary language and the other very much less so. Knowledge of a value concept is something to be understood, as it were, in depth and not in terms of switching on to some given impersonal network.… We do not simply, through being rational and knowing ordinary language, "know" the meaning of all necessary moral words. We may have to learn the meaning; and since we are human historical individuals the movement of understanding is onward into increased privacy, in the direction of the ideal limit, and not back towards a genesis in the rulings of an impersonal public language' (Murdoch, 1985(Murdoch, [1970: 29). 6. Concepts that are conceived by some philosophers as having a descriptive and an evaluative part are called 'thick' moral concepts (e.g. Williams, 2006Williams, [1985: 129ff.). The idea that only some concepts are like that has been used in the context of moral philosophy to delimit its subject matter by attempting to identify a distinctly 'moral vocabulary'. The Wittgensteinian tradition Winch is part of, however, does not find any sense in drawing a limit on what expressions can receive moral significance, as Duncan Richter (2000: 45) points out with the use of a powerful example from an unpublished paper by Cora Diamond (1991: 9ff.), who herself illuminatingly discusses ethics as not a particular subject matter but an attitude, a dimension that may crop up. But see De Mesel (2015). 7. I have argued elsewhere that the appeal to art is not something that can be turned into a methodological recommendation for social science (Tsilipakos, 2018). As to the need to resort to extraordinary language, this is why Gaita seems to think, and possibly Winch in ISS, that Ordinary Language Philosophy (OLP), with its 'what we would say whens', does not quite deliver. For what it is worth, I do not share Gaita's views. I think he is being unfair to OLP in, first, misconstruing what 'ordinary' is actually contrasted with, second, overlooking the emphasis it places on points of logic in an extended sense, but not in terms of settling between competing normative conceptions and, third, underestimating the sensitivity required to show the differences between concepts, such as being just, delicious or green, to refer to Pitkin's (1972) examples, for instance, in terms of the kinds of questions and challenges that can be raised and justifications offered: 'You call that justice, delicious, green?', 'How is it justice when those people are wronged, but it completely lacks salt, can't you see those red hues?'. 8. There is a differencenot always sharp, but a difference nonethelessbetween being able to register a moral vision and being oneself committed to it. The idea that personal commitment may be necessary for understanding does indeed apply to some relatively extreme cases but cannot efface the meaningfulness of those distinctions. Winch himself was concerned with precisely the kinds of limit to understanding that are to be found even within our own culture in his 'Can We Understand Ourselves' (Winch, 1997). Yet one should bear in mind that understanding is not either entirely present or entirely absent but rather something relative to various purposes. One can understand something well enough to consider it or to talk about it or to want to learn more about it, to convey a sense of it to a third party, or to cite the words of those who participate in the activity with some understanding without necessarily subscribing to it or adopting the exact same perspective. It is not correct either to say that this is never possible or to say that this is always possibleconveying what one considers an unthinkable position in intelligible terms may on occasion not be possible, but this is hardly something general enough to be considered a governing principle of social science. 9. A. R. Louch, who initially reviewed ISS (Louch, 1963(Louch, , 1965 and engaged in a critical exchange with Winch (1964a), puts forth a thesis in his own monograph that is arguably very close in spirit to Winch's. He wonders how human action is to be described if not in the morally sensitive ways that open it up to detail and appraisal and proceeds to claim that 'what is needed, in both psychology and ethics, is not measurement, experiment, prediction and formal argument, but appraisal, detailed description, reflection and rhetoric.… Psychology and social science are moral sciences; ethics and the study of human action are one' (Louch, 1966: 234-5). There is some irony in the fact that the same scholar who criticised Winch for running philosophy and the social sciences together (ibid.: 174ff.) so bluntly asserts the unity of philosophical ethics and the study of human action. In that respect, Louch and Winch are mostly in agreement. 10. The culprit might be identified as lack of alertness about the limits of some of Wittgenstein's ways of speaking. But it may be worth noting as a matter of scholarly interest that some of Winch's formulations in question are also very close to Michael Oakeshott's. Consider what Oakeshott says in his 1950 essay 'Rational Conduct': 'An activity as a whole (science, cooking, historical investigation, politics or poetry) cannot be said either to be "rational" or "irrational" unless we conceive all idioms of activity to be embraced in a single universe of activity' (Oakeshott, 1991: 122). Finally, we might include another observation made in Experience and Its Modes, where Oakeshott (1985Oakeshott ( [1933: 334) discusses ethical thought as 'incapable of constituting itself as a self-determined, homogeneous world of its own'. 11. Winch (1990b) refers to Rhees, who emphasises precisely the range of interrelated activities that are necessary in order to give 'an idea of a people with a definite sort of life. Do they have songs and dances and festivals, and do they have legends and stories? Are they horrified by certain sorts of crimes, and do they expose people to public ridicule?' (Rhees, 1959: 184;cf. Mounce and Phillips, 1969: 77). Winch also took up the point throughout his own work, emphasising the 'multiplicity of [institutions, traditions, and forms of life] which interpenetrate each other both in society at large and also in the lives of particular individuals. This means that there is genuine room for individual decisions in questions of social policy: decisions about where one's allegiances lie, what considerations one is going to take as most important, and so on' (Winch, 1974: 900;original emphasis). For another example, regarding science and religion, Winch stresses in 'Darwin, Genesis and Contradiction' that they are not isolated but 'untidily tangled together, they react on each other, and, what is more, they may seem sometime to come into conflict' (Winch, 1987: 132). It is possible to maintain a life of coherence but not because these activities are to be kept separate: a person is vulnerable to criticism if they go out of their way to prevent 'the mingling of incense and nitrogen sulphide', as Winch nicely phrases matters in 'Meaning and Religious Language' (ibid.: 121).