Engineering is not a luxury: Black feminists and logical positivists on conceptual engineering

ABSTRACT Recent historical discussion of conceptual engineering by analytic philosophers has largely focused on precedents for contemporary conceptual engineering within the history of analytic philosophy. However, I suggest that we can and should look outside of the analytic tradition for further examples of conceptual engineering, and inspiration for further work in conceptual engineering. Here I will look to one such other tradition – American Black feminism. I do this by considering the work of Audre Lorde and Patricia Hill Collins in tandem with a tradition that is the more usual stomping ground of analytic philosophers: logical positivism. I draw out Rudolf Carnap and Otto Neurath’s respective views on conceptual engineering, before turning to Collins and Lorde. I suggest that Collins on the power of self-definition provides a model of conceptual engineering that closely matches that given by Neurath, whilst we can read Lorde’s work on poetry as giving a distinctively individualist spin on the conceptual engineering. I conclude with a comparative discussion of the various models on offer, suggesting that rather than being in competition, these various views of conceptual engineering are best seen as a toolbox of methods for conceptual engineering.


Introduction
Recent historical discussion of conceptual engineering by analytic philosophers has largely focused on precedents for contemporary conceptual engineering within the history of analytic philosophy. 1 However, I suggest that we can and should look outside of the analytic tradition for further examples of conceptual engineering, and inspiration for further work in conceptual engineering. Here I will look to one such other tradition -American Black feminism. I do this by considering the work of Audre Lorde and Patricia Hill Collins in tandem with a tradition that is the more usual stomping ground of analytic philosophers: logical positivism. Given the apolitical, if not conservative reputation of the logical positivists, this might seem an odd pairing, thus I will begin by dismissing concerns that the positivists are ill-suited to thinking about the sorts of emancipatory projects that concern many Black feminists. I then draw out Rudolf Carnap and Otto Neurath's respective views on conceptual engineering, before turning to two black feminists who, I argue, we can read as giving different visions of conceptual engineering -Patricia Hill Collins and Audre Lorde. I suggest that Collins on the power of self-definition provides a model of conceptual engineering that closely matches that given by Neurath, whilst we can read Lorde's work on poetry as giving a distinctively individualist spin on the conceptual engineering. I conclude with a comparative discussion of the various models on offer, suggesting that rather than being in competition, these various views of conceptual engineering are best seen as a toolbox of methods for conceptual engineering.

Logical positivists: some stereotypes repudiated
Stereotypes persist regarding the logical positivists that they were apolitical, dismissive of ethics, and can basically be seen as fusty bureaucrats interested only in logic, mathematics, and the philosophy of science. Meanwhile, positivism more broadly is seen as conservative, hiding behind a veneer of political neutrality. I take it that these stereotypes suggest that the logical positivists are ill-suited to engaging in emancipatory projects, 2 and, if anything, would suggest that a logical positivist would be opposed to the projects of say, trans liberation, anticapitalist class warfare, or anti-racist struggle. It looks, therefore, like the project of drawing comparisons between Black feminists and logical positivists is off to a rocky start. If the stereotypes are to be believed, we have two deeply opposed sets of thinkers.
However, I will suggest that even if we cannot say for sure that Carnap and Neurath would have embraced Black feminism in some form, they were nonetheless committed to emancipatory projects writ broad, and may well have been sympathetic to Black feminist goals and activism. In the rest of this section, I will try and spell out some aspects of that commitment by drawing on both philosophical and biographical details about Carnap and Neurath. Neurath and Carnap were committed the scientific world conceptionan empiricist, positivist outlook that suggested that science, with the aid of logical analysis, can provide a unified picture of the natural and social worlds. This world conception is tied to a commitment to scientific humanismthe thought that emancipation can best be achieved only with the aid of science. The thought here is, as Carnap put it, 'the scientific method is the best method of acquiring knowledge and that therefore science must be regarded as one of the most valuable instruments for the improvement of life' (Carnap 1963, 83). This is contrasted with what they took to be the metaphysical outlooks provided by the likes of their contemporary Heidegger, which they took to be tied to forms of fascism. Notably, in their manifesto for the Vienna Circle, Neurath and Carnap, along with Hans Hahn, express this position as follows: Thus the scientific world-conception is close to the life of the present. Certainly it is threatened with hard struggles and hostility. Nevertheless there are many who do not despair but, in view of the present sociological situation, look forward with hope to the course of events to come. Of course, not every single adherent of the scientific world-conception will be a fighter. Some, glad of solitude, will retreat to the icy slopes of logic … However, their achievements too will take a place among the historic developments. We witness the spirit of the scientific world-conception penetrating in growing measure the forms of personal and public life, in education, upbringing, architecture, and the shaping of economic and social life according to rational principles. The Scientific World-Conception serves life, and life receives it. (Carnap, Hahn, andNeurath 1929/1973, 317-318) Neurath was perhaps the most openly political of the members of the Vienna Circle. Prior to joining the Circle, Neurath was head of the central planning office of both of the short-lived Bavarian Soviet Republics. He advocated a radical plan of full socialisation of the economy, convinced that the principles of war economy could be translated into an economy aimed at serving the happiness of people (Neurath, quoted in Cartwright et al. 1996, 44). Upon the collapse of the republic at the hands of government forces Neurath was convicted of assisting high treason but was deported to Vienna before he could serve his full sentence. In Austria he took part in the milieu of Red Vienna, where he became an active participant in the housing movement, campaigning for public housing and allotments. This political radicalism was bound up with pedagogy: he not only founded museums for the education of workers, but also developed Isotypea visual language for the communication of science.
Carnap, meanwhile, joined the Independent Socialist Party in 1917, and with Karl Bittel began an underground journal Political Curricular. Like the other members of the Vienna Circle who had not already been killed, he had to flee Vienna upon the rise of fascism. He arrived in the US and became involved in political activities there, writing critically of US government, especially over foreign policy. Throughout his life he remained internationalist and expressed sympathy for world-government and central planning. Even his interest in Esperanto seems to have been motivated by this internationalism. This political tendency was somewhat dampened by McCarthyism however: As new immigrants, Carnap and his Vienna Circle friends felt vulnerable in the United States, and many of them, having previously supported socialist causes in Europe, were kept under observation by the FBI, including Carnap himself. It was thus in their interest to play down the larger significance of their philosophical preoccupations, and to pose as narrow specialists in technical subjects. (Leitgeb andCarus 2020, see also Reisch 2005) This, we might speculate, is perhaps what has left us with the stereotypes mentioned above. This said, Carnap did not give up on his political activities entirely. The FBI, for instance, notes that Carnap signed a petition that appeared in the east-coast communist newspaper The Daily Worker, and their files contain over 60 pages details the more 'political' side of Carnap's life, including such shocking details as that he endorsed a pamphlet calling for world peace. 3 I take it that the brief discussion above is enough to demonstrate that Carnap and Neurath had a commitment to emancipatory projects broadly 3 A (perhaps apocryphal -Putnam couldn't remember the source) remark from Hilary Putnam on Carnap's response to McCarthyism is recounted by Alan Gilbert: 'Carnap rightly felt that the only decent response to McCarthyism was to join the Communist Party. He tried to look it up in the Los Angeles phone book, but the Party, under attack, had gone underground and was no longer was listed. So he couldn't join'. (Gilbert 2016). Supposedly Carnap just gave up on joining thereafter! construed, with a particular emphasis on economic justice, but can we say anything more that might suggest a sympathy for black feminism? Well, whilst we perhaps cannot link them to black feminism directly, we can certainly appeal to the consistent anti-racism of the logical positivists.
As Liam Bright has shown, they were often 'overtly hostile to racial explanations' (Bright 2017, 12) and 'frequently expressed hostility to ethnic or racial prejudice, and support for political acts which promoted more equitable racial relations' (Bright 2017, 13). In particular, Bright notes that in Neurath we see debunkings of stereotypes about Asian people (Neurath 2017), and mockery of Nazi race theory (Neurath 1946). Moreover, we can point to Carnap's work in Los Angeles, where he was involved in a number of civil rights organisations, including Operation Bootstrap, which was an offshoot of the Congress of Racial Equality (Sarkar 1992, 7). Where does this leave us? Well, we can say the following: Carnap and Neurath were committed to a number of emancipatory projects. Far from the stereotype of politically neutral or indeed conservative positivists, Carnap and Neurath demonstrated a political radicalism and importantly a commitment to racial justice. This is far from demonstrative proof that they would have been on board with the Black feminisms of Collins or Lorde, but I want to suggest that the above at least makes it seem reasonable to think that we might gain much from a comparison of the figures, or at least that their work is commensurable. 4 Below, I turn in particular to these thinkers' work on conceptual engineering. Here I suggest that we get more than mere commensurability.

Conceptual engineering in logical positivism
Conceptual engineering is not new, but it has seen a recent revival in analytic philosophy following the work of Sally Haslanger (see Haslanger 2012). Like a lot of things, it can be difficult to define, but, broadly construed, we might think of it as critical reflection on our concepts, replacing them with improved concepts that suit our purposes better. But note that it need not necessarily concepts that are at issue here (see Capellen 2018). Depending on one's stance on the nature of concepts, one could be interested instead merely in engineering word meanings, linguistic forms/practices, logical systems and so on. As such, conceptual engineering is probably best generally characterised as the improvement of our representational devices (whatever those are) in order to better suit our purposes.
Perhaps the best-known engineer of the early twentieth century was indeed Rudolf Carnap. For Carnap, the project of providing and exploring exact languages for usage in science and mathematics becomes a project in making precise those vague concepts already being used, in what he calls explication. 5 He writes that 'By the procedure of explication, we mean the transformation of an inexact, prescientific concept, the explicandum, into a new exact concept, the explicatum.' (Carnap 1962, 3). The thought here is that we need precise concepts for science, but ordinary language is vague and often obfuscatory. Thus, says Carnap, we ought to update our concepts in ways that are useful to science. In order to achieve this end, we need some way of evaluating our putative improved versions of explicanda. To this end Carnap suggests four desiderata: Similarity to the explicandum: 'in most cases in which the explicandum has so far been used, the explicatum can be used; however, close similarity is not required, and considerable differences are permitted'. (Carnap 1962, 7) Exactness of explicatum: 'the rule of its use (for instance in the form of a definition), is to be given in an exact form, so as to introduce the explicatum into a well-connected system of scientific concepts'. (Carnap 1962, 7) Fruitfulness of explicatum: the explicatum must be 'useful for the formulation of many universal statements (empirical laws in the case of a nonlogical concept, logical theorems in the case of a logical concept)'. (Carnap 1962, 7) Simplicity of explicatum: 'easured, in the first place, by the simplicity of the form of its definition and, second, by the simplicity of the forms of the laws connecting it with other concepts'. (Carnap 1962, 7) One thing to note here is that these are all epistemic desiderata. No moral or pragmatic concerns enter directly in the practice of conceptual engineering for Carnap. Of course, given his broader scientific humanist commitments, we might think that moral and political considerations motivate Carnap's project more broadly, but such values do not come into the practice of engineering concepts. One way of putting this is to claim that moral and political values are achieved in Carnap's envisaged project downstream of the practice of explication. 6 This is basically the idea that in doing science well (and in particular in doing explication well) one need not incorporate moral and political values directly into one's methodology, but that doing science well in this way has consequences of moral and political value, by enriching the lives of ordinary people.
It is on this point that Carnap was challenged by Neurath, who in my terms suggested to Carnap that non-epistemic values ought to enter into the practice of explication upstream, helping us decide on which concepts we ought to adopt: The choice of a language form was a practical decision, [Neurath] argued, just as the choice of a route for a railroad, or that of a constitution for a government. He emphasized that all practical decisions are interconnected and should therefore be made from the point of view of a general goal. The decisive criterion would be how well a certain language form, or a railroad, or a constitution, could be expected to serve the community which intended to use it. (Carnap 1963, 51) 7 As presented by Carnap, Neurath recommends adopting different languages to suit the purposes of different communities, where (given the analogies of railroads and constitutions) these purposes may be practical, political, and may take other forms as well, so long as they are the purposes of the community being served by the language. Thus, whilst Carnap's methodology for explication involves merely epistemic values entering to our concept-engineering procedure, Neurath's methodology [Neurath's] emphasis on the interdependence of all decisions, including those in theoretical fields, and his warning against isolating the deliberation of any practical suggestion, even that of the choice of a language form, made a strong impression on my own thinking and that of my friends. (Carnap 1963, 51) However, 'having a strong impression' and 'persuading of the fact that' are two different things. An interesting interpretative wrinkle here is that this comment occurs in Carnap's discussion of physicalistic language, and Carnap suggests, immediately following this comment, that one advantage of the 'physicalistic language is its intersubjectivity, i.e., the fact that the events described in this language are in principle observable by all users of the language' (Carnap 1963, 52). One piece of evidence that would push us to think that Neurath persuaded Carnap on the place of nonepistemic values in explication, then, would be if we could show that Carnap thought of intersubjectivity as a primarily pragmatic or moral/political value. Whilst a full discussion of this issue goes beyond the scope of this paper, we might think that an argument for this claim drawing on broadly scientific humanist claims is at least prima facie plausible.
is pluralist about the kinds of values that may enter. 8 Neurath's methodology is also pluralist in another sense: given that different communities will have different ends, it seems plausible that in at least some circumstances the concepts that the different should adopt will differ. We might think, for instance, that a group of logicians engineering a concept of truth for usage as a part of formal inquiry might plausibly want a different concept of truth from a community of teachers attempting to teach small children not to lie.
I think we can reconstruct Neurath's position as follows, where the community whose purposes are guiding the ameliorative inquiry is C p , and the community of intended users of the concept is C u : Stage One: Identify a concept for engineering T and set out the purposes P for which a community C p seeks a concept, concepts, or lack thereof.
Stage Two: Determine the best concept or concepts to use, or whether to give up on all the concepts for T given P and the intended community of users of the concept(s) C u .
Stage Three: Encourage the adoption of the new concept(s) or the adoption of abolition about concepts for T in C u , given P. (Cull 2020, 18) 9 Carnap and Neurath thus offer us two different methodologies for conceptual engineering. The former suggests that explication is undertaken by scientists, philosophers, and logicians. Moreover, Carnap's explication is only indirectly political, allowing only epistemic values into the development of new concepts, and due to his scientific humanist commitments, Carnap trusts that this epistemically driven engineering will bring about human liberation and increases in happiness. Meanwhile the latter offers a version of conceptual engineering wherein political, practical, and other values enter into the development of new concepts by communities in the light of those communities' interests. We will return to these positions on conceptual engineering later in the paper. For now, though, we turn to Black feminism.

Conceptual engineering in Black feminism
Whilst contemporary conceptual engineers have engaged with the logical positivist tradition in thinking through methodological questions, historians of conceptual engineering have yet to examine the conceptual engineering projects undertaken by Black feminists in the twentieth century. I want to suggest that Patricia Hill Collins and Audre Lorde offer distinctive views of conceptual engineering, even if they do not label their ideas as such. Collins, I will suggest, offers a community-based view that has much in common with the view put forward by Neurath, whilst Lorde takes a more individualist position. Importantly for both, what I am calling their versions of 'conceptual engineering' play an important role in their political praxis as Black feminists. As I shall detail, they see 'conceptual engineering' as playing an important role in resisting and overthrowing oppression.

Collins on controlling images and the power of self-Definition
Collins suggests that black feminists are concerned to resist 'controlling images'. That is, stereotypes, symbols, and ideas of black womanhood that are used as a means of dominating black women: These controlling images are designed to make racism, sexism, poverty, and other forms of social injustice appear to be natural, normal, and inevitable parts of everyday life. (Collins 2009, 76-77) Some examples Collins considers include the image of the 'mammy', the 'matriarch', and the 'welfare mother'. She rightly points out that such images reinforce racist and misogynist hierarchies and define what it means to be a black woman in the terms of white men. For our purposes here, however, our interests lie in how she describes Black women's responses to these controlling images.
Collins suggests that there are a number of ways that black women thinkers and writers have responded to controlling images, but a key way in which controlling images can be resisted comes from selfdefinition within safe spaces. She suggests that places like extended families, churches, African American Community organisations, and even the blues musical tradition provide space for reflection and resistance away from hegemonic white culture. Quoting Sondra O'Neale, she writes: These spaces are not only safethey form prime locations for resisting objectification as the Other. In these spaces Black Women 'observe the feminine images of the 'larger' culture, realize that these models are at best unsuitable and at worst destructive to them, and go about the business of fashioning themselves'. (Collins 2009, 111) Thus, safe spaces are places where Black women can come together to form a critical consciousness of the way in which they are represented by white supremacy, and moreover, develop new ways of representing themselves in ways that are not distorted by white supremacist ideology: These sites offered safe spaces that nurtured the everyday and specialized thought of African-American women. In them Black women intellectuals could construct ideas and experiences that infused daily life with new meaning. These new meanings offered African American women potentially powerful tools to resist the controlling images of Black womanhood. Far from being a secondary concern in bringing about social change, challenging controlling images and replacing them with a Black women's standpoint constituted an essential component in resisting intersecting oppressions. (Collins 2009, 123 My Emphasis) Collins goes through a variety of self-definitions that have arisen from safe spaces. Notable features include those of self-valuation, self-reliance, selfsufficiency, and autonomy, but also community healing, respect, and connectedness among individuals (Collins 2009, 123-130). Important to all, however, is the rejection of the images imposed by a racist and misogynist society. A rejection of these images, suggests Collins, is an important precondition of liberation from misogynoir.
I want to suggest that whilst she does not use the term itself, that nonetheless Collins is recommending a particular form of conceptual engineering here. In particular, I suggest that she is recommending a form of conceptual engineering that is based around communities in safe spaces, new concepts of black womanhood are developed that aid in resisting oppression.
How do these new concepts aid in resisting oppression? Following Lorde, Collins writes, By struggling for self-defined womanist perspectives that reject the 'master's' images, African-American women change ourselves. A critical mass of individuals with a changed consciousness can in turn foster Black women's collective empowerment. A changed consciousness encourages people to change the conditions of their lives. (Collins 2009, 129). 10 As far as I can tell, Collins suggests that we see at least three ways that Black women are aided in resisting oppression by this version of conceptual engineering. The first and perhaps most obvious is a causal forcethat equipped with these new ways of thinking about themselves and their relationship to an oppressive society, Black women are motivated to make collective change, and are equipped with better conceptual tools to make that change. The second is personal, the thought being that equipped with new concepts of Black womanhood, Black women experience a sense of personal empowerment, 'even within conditions that severely limit one's ability to act' (Collins 2009, 129). The third is constitutivethat given new concepts, social realities can be redefined (Collins 2009, 130). 11 Nicely, I think that Collins' ideas here generalise to other oppressed groups. We might, for instance, think that disabled people, or trans people, can, by forming, or using already-existing safe spaces, develop new meanings and concepts that help resist the harmful effects of ableist or transphobic ideology. Indeed, it seems plausible to me that such conceptual engineering is already occurring and has been occurring for some time. For instance, we might read the development of the social model of disability by the Union of the Physically Impaired Against Segregation (UPIAS) as a development of new concept of disability from within a safe space. Notably UPIAS held both closed meetings and had a private circular in which ideas and 'a new concept' (Hunt 2001) of disability could be developed, by disabled people, away from ableist society. Meanwhile, Cull (2020) has suggested that trans people avail themselves of the resources of conceptual engineering in developing new gender concepts, tying in with what Ashley has called an 'ethics of gender exploration' (Ashley 2019). We might think that in online trans chatrooms, and in-person organising meetings trans people precisely step back from cissexist society and into safe spaces to use what Collins calls the power of self-definition.

Lorde's poetry as engineering
Similarly, I suggest that we can read Audre Lorde as putting forward a method of conceptual engineering via poetry. There are several places 11 For more on causal versus constitutive construction, see Haslanger 2012. I will note that I am somewhat sceptical of Collins' third claim here. It is hard to see, for instance, how social reality could change just in virtue of new definitions, absent material changes, whether those be social action, economic disruption, or environmental change. Discussion of this question goes well beyond the scope of this paper, though see Marx 1867/1976 for a starting point on this literature.
in her essays where Lorde signals her concern with our representational devices, and engineering new ones in places where individuals face hermeneutical lacunae, most notably in her essay 'The Transformation of Silence into Language and Action', where she can be read as suggesting that black women (and other oppressed groups) need new concepts in order to better understand their experience and make change: What are the words you do not yet have? What do you need to say? What are the tyrannies you swallow day by day and attempt to make your own, until you will sicken and die of them, still in silence? (Lorde 2017, 3) Here Lorde suggests that finding the right words to express oneself is a perhaps necessary precondition of overcoming oppression, the thought being that we must be able to elucidate our problems in order to solve them. This idea of clarification of issues through the clarification of our representational devices to help us come up with solutions to the problems we face, is expanded upon in her 'Poetry is not a Luxury': The quality of light by which we scrutinize our lives has direct bearing upon the product which we live, and upon the changes which we hope to bring about through those lives. It is within this light that we form those Ideas by which we pursue our magic and make it realized. This is poetry as illumination, for it is through poetry that we give name to those ideas which are, until the poem, nameless and formless-about to be birthed, but already felt. That distillation of experience from which true poetry springs births thought as dream births concept, as feeling births idea, as knowledge births (precedes) understanding. (Lorde 2017, 7) Here I wish to suggest that we should not be too narrow in our definition of 'poetry'. One reading of the above would be to suggest that new concepts arise specifically from poetry, where poetry is a narrowly defined type of literature, perhaps featuring rhyme, metre, and so on. However, I want to suggest that Lorde instead has in mind 'poetry' broadly defined as any work that closely describes one's experience: I speak here of poetry as the revelation or distillation of experience, not the sterile word play that, too often, the white fathers distorted the word poetry to mean … Poetry is the way we help give name to the nameless so it can be thought. The farthest horizons of our hopes and fears are cobbled by our poems, carved from the rock experiences of our daily lives. (Lorde 2017, 8) By 'poetry', Lorde means more than just verse! Instead, poetry is a broad category of autobiographical descriptive work that closely engages and describes our experiences. 12 This work, she suggests, generates new words, new ideas, and new concepts, which can then be put to work in fighting oppression. At this point one might be tempted to ask, what is it about the close description of experience that throws up these new concepts? Lorde, I suspect, might answer in one of two ways. The first is simply to suggest that the work of close description and thinking about one's experience is just the sort of thing that makes us re-think terminology and forces us to come up with new or modified ways of describing difficult to explain features of our world. In more mystical moods, however, Lorde sometimes suggests that there is a 'deep place within' that is the source of creativity and indeed new concepts, which are teased out through the process of close descriptive autobiographical work. This alternative is suggested in her poem 'Black Mother Woman', which she quotes as a part of 'Poetry is not a Luxury': For each of us as women, there is a dark place within where hidden and growing our true spirit rises, 'beautiful/and tough as chestnut/stanchion against your nightmare of weakness' … These places of possibility within ourselves are dark because they are ancient and hidden; they have survived and grown strong through that darkness. Within these deep places, each one of us holds an incredible reserve of creativity and power, of unexamined and unrecorded emotion and feeling. (Lorde 2017, 7-8) The thought here is that we only need to tap into this creative resource, examine its contents, and we can get new concepts out of it. Sara Ahmed seems, in her interpretation of Lorde's The Cancer Journals, to endorse something like the former, less mystical reading, suggesting that it is the description of a particular experience (the loss of her right breast to a mastectomy) that leads to the production of new concepts. Whilst Ahmed does not make an explicit link to the analytic tradition of conceptual engineering, I nonetheless take her to be implicitly agreeing with my suggestion of reading Lorde as a conceptual engineer when she, Ahmed, argues that we can read Audre Lorde as producing new, sweaty concepts: When I use the concept of 'sweaty concepts' I am also trying to say we can generate new understandings by describing the difficulty of inhabiting a body that is not at home in a world. Sweat is bodily; we might sweat more during more strenuous activity. A 'sweaty concept' might be one that comes out of a bodily experience that is difficult, one that is 'trying,' and where the aim is to keep exploring and exposing this difficulty. (Ahmed 2014) For Ahmed, Lorde is describing how to produce new concepts from a position of not being at home in the world, or, to put it another way, a position of oppression. These new concepts take a particular form for Ahmed they were hard work to produce, but can do work for us in making political change. I am hesitant to suggest that the development of new concepts via close descriptive work will always be difficult or sweaty, after all, I am sure that many readers will be familiar with the feeling that sometimes it can be a great relief to just 'get it all out'. 13 That said, I think that Ahmed captures something quite important about the kind of political purposes that can enter into Lorde's methodologywe are often developing new concepts that we want to do work for us in opposing oppression.
This close connection between poetry, newly engineered concepts and political activism leads Lorde to a great sense of urgency when discussing poetrythat rather than being a luxury, poetry is a political necessity given the ongoing oppression of Black women, and indeed other groups. This comes to a head in a 1984 conversation between Lorde and James Baldwin: AL: I can't tell you what I wished you would be doing. I can't redefine masculinity. I can't redefine Black masculinity certainly. I am in the business of redefining Black womanness. You are in the business of redefining Black masculinity. And I'm saying, 'Hey, please go on doing it,' because I don't know how much longer I can hold this fort, and I really feel that Black women are holding it and we're beginning to hold it in ways that are making this dialogue less possible.
And later in that same conversation: AL: We have to begin to redefine the terms of what woman is, what man is, how we relate to each other.

JB:
But that demands redefining the terms of the western world … AL: And both of us have to do it; both of us have to do it … (Baldwin andLorde 1984/2014). 14 What is the relationship between Collins' version of conceptual engineering, as presented above, and Lorde's version? While both are concerned with the development of concepts for the liberation of Black women, Collins' methodology places a greater emphasis on the community than Lorde's. Lorde suggests that the source of conceptual engineering is not the community, but the individual, describing their experience closely. Of course, this is not complete isolation -Lorde's methodology importantly relies on the close description of one's social experience, and one's perspective that is socially located in a particular waybut certainly, the focus is not on collective development of concepts.

Comparison of the methodologies
We now have four different models of conceptual engineering on the I think this points to a close affinity between Collins and Neurath on conceptual engineering. Both look to communities to develop new concepts in line with the demands of those very communities. Indeed, due to this focus on communities, both are committed to a form of pluralism about the concepts which should be adopted: different communities are going to produce different concepts to fulfil different demands. Moreover, we might think that anyone attempting to do Neurathian conceptual engineering well is basically going to be doing Collinsian conceptual engineering. That is, they are going to suggest that it is safe spaces that provide the best place to develop new and improved concepts. At least with regards to concepts for the purposes of oppressed groups, sharing the demands of communities, and developing new or modified concepts in light of those demands looks like it is going to be best accomplished in safe spaces for the reasons that Collins suggests. We might think that the same conclusion will also hold true for communities of scientists for particular subfields, however. Whilst the safe spaces may look slightly different (specialist conferences rather than consciousness raising groups, peer-reviewed journals rather than small-circulation zines) we might nonetheless think that these scientific safe spaces are the best places for developing new and improved concepts for communities of scientists.
We might also think that there are some similarities between Lorde and Carnap. For one, it looks as if Carnap has a conception of poetry that shares much with the conception of poetry I suggested that Lorde puts forward. For Carnap, poetry, along with other art forms (especially music) provide us with a means of giving expression to one's basic attitude to lifeone's 'emotional and volitional reaction to the environment, to society, to the tasks to which he devotes himself, to the misfortunes that befall him' (Carnap 1959, 79). This should strike us as a remarkably familiar idea of the role of poetry, one that matches closely to the vision I suggested that Lorde offered above. This said, Carnap's remarks on poetry come in his famous attack on metaphysics as meaningless, and his assertion that metaphysicians are artists without talent. We might, therefore, suggest that Carnap would be sceptical that poetry can provide the kinds of cognitively significant resources that Lorde suggests it can. After all, even if poets have talents with regards to the expression of basic attitudes, they are for Carnap doing something analogous to Metaphysicians, and will not be able to produce theoretically significant or meaningful contentthey will be firmly 'in the domain of art and not in the domain of theory' (Carnap 1959, 80). This said, I think that we can instead hold that it is consistent with Carnap's position that even if poetry itself does not involve cognitively significant or meaningful content, nonetheless poetry can give rise to concepts that are useful for our various projects. That is, the expression of a basic attitude via poetry might inspire us to re-think our language and concept choices. 15 15 Given Carnap's approving remark on Dilthey and his followers here (Carnap 1959,79) we might think that drawing on Dilthey offers a way of thinking about how the Carnapian might suggest that these new concepts might be drawn directly from art. As Philipson puts it, 'Dilthey's point concerning the value of a work of art is not that it competes with science in contributing to knowledge, but that it offers another mode of apprehending the ultimate object and the ultimate subject: life'. (Philipson 1958, 73). We might think that an alternative mode of apprehension is precisely the type of thing that would provide the kinds of creative friction useful for inventing new and improved concepts. Thanks to an anonymous reviewer for pushing me to engage with Carnap on poetry.
More tightly linking Lorde and Carnap on conceptual engineering is the thought that if we think of Lorde as suggesting that we ought to undertake conceptual engineering qua poetry in order to produce new concepts that can serve as tools in fighting oppression, then there seems to be at least some overlap with Carnap's model, even if Carnap thinks that the tools ought to be aimed at improving science, at least in the first instance. After all, if, as Lorde herself famously puts it, 'the master's tools will never dismantle the master's house' (Lorde 2017 , 91), it looks as if it is in conceptual engineering as poetry, we produce the new tools that enable us to burn down that house.
There is a worry hiding in the background here, however. Even if conceptual engineering can be used in this way, to produce new concepts that can be used to fight oppression, does the above discussion of conceptual engineering as practiced by logical positivists not imply that conceptual engineering one of the master's tools? After all, aren't the logical positivists the paradigm case of the developer of the master's tools? Even if conceptual engineering does not originate with them, they certainly embrace it. If this is so, then we might be worried that there is a tension between a rejection of the master's tools and an embrace of conceptual engineering in the work of the Black feminists we have been looking at.
One immediate response to this problem would be to suggest that conceptual engineering is not one of the master's tools precisely because it enables us to rethink the very concepts which make up the master's tools. Indeed, is this not precisely why Lorde adopts her version of conceptual engineering in the first place? However, I think this is a little too quick. If we are taking the master's tools slogan seriously and are interested in radically critiquing all of our methods and conceptual apparatuses on such grounds, we need some reason to think that conceptual engineering itself is exempt from such criticism.
Another response might be to suggest that conceptual engineering in the hands of Lorde and Collins is radically different from conceptual engineering in the hands of Carnap and Neurath. However, I am sceptical that this response is going to workas argued above, Collins and Neurath offer extremely similar visions of conceptual engineering, and we can even point to some similarities between Lorde and Carnap's respective methodologies. One could, of course, stress the distinctive features of Lorde's accountclose autobiographical description and a greater degree of individualism than is present in other versions of conceptual engineering, but a story would need to be told about what it is about these features of Lorde's methodology that stops it from being one of the master's tools.
Instead, I suggest that our response to this problem should be to deny that conceptual engineering, as practiced by the logical positivists, is one of the master's tools. As argued at the start of this paper, the logical positivists, and Carnap and Neurath in particular, were radicals whose political goals included fighting fascism, implementing a more economically just society, and a commitment to racial equality, if not Black liberation. If this is the case, then I suggest that it becomes plausible that this tradition of conceptual engineering just is not a master's tool. Rather, it is a tool that was developed by people trying to tear down the master's house. 16 Black feminists, then, I suggest do not face a tension between a commitment to Lorde's slogan about the master's tools and the employment of conceptual engineering. 17

On the uses of engineering
Of course, nothing I have said above should be taken as an argument for the superiority of one thinker's methodology for conceptual engineering over and above any others. I have not argued for the superiority of any one method from the various accounts that we have seen, and this is quite on purpose. Instead, I suspect that the various methods will be found amenable to different circumstances, and different purposes. In clarifying the methods laid out by Carnap, Neurath, Collins, and Lorde, I hope to have provided something like a menu of options for potential conceptual engineers. I take a lesson from the above to be that there are a number of methods available, and that we should select the particular method we will want to use on the basis of our particular goals. 16 Audrey Yap argues for a similar conclusion regarding the compatibility between the 'master's tools … ' slogan and Carnapian methodology, though her concern there is largely logic rather than conceptual engineering, and she reaches her conclusion via a different argument. Rather than offer a genealogy that suggests that the methodology was never one of the master's tools in the first place, she suggests that logic, when undertaken in the right manner, need not be one of the master's tools (Yap 2010, 453n6). 17 An anonymous reviewer points out that one might read the above and suggest that nonetheless even if Carnap and Neurath did aim their tools to be ones that were useful in bringing down the master's house, that does not imply that they were successful in their aim. Perhaps they accidentally ended up developing and using one of the master's tools in spite of themselves. This is, however, a worry not just for Carnap and Neurath, or even just for conceptual engineering, but rather any putatively emancipatory strategy or concept. The effects of ideology are plausibly so pernicious as to obscure what is actually a radical break from the master's toolssome strategy might look revolutionary at first glance, but ultimately end up reinforcing the very oppressive structures it was designed to oppose. This is a deep issue which goes well beyond the scope of this paper to address, but see the literature on ideology critique, especially Haslanger (2017).
Think, for instance, of two different casesthe first an isolated trans person coming to terms with their gender and trying to figure out the best way to make sense of that gender in an environment that is hostile to them. Perhaps due to geographical fluke they are unable to access the resources often provided by trans communities in large cities and are largely forced to deal with their identity in isolation. Here we might think that a Neurathian/Collinsian version of conceptual engineering is poorly suited, if only because this person lacks a community with whom to do it. However, Lorde's methodology might seem apt: it might be suggested that via diary, narrative, or other close description of their experience, the trans person might come up develop or modify concepts in such a way as to best make sense of their experience. Indeed, something like such a practice has been documented by Singh et al, who note that journaling, for instance, was used successfully by participants 'to explore defining their gender, identifying the specific words and related meanings they used to define their gender' (Singh, Meng, and Hansen 2014, 211).
The second case is that of a group of feminists trying to make sense of repeated unwanted sexual behaviour at work, as described by Brownmiller: The 'this' they were going to break the silence about had no name. 'Eight of us were sitting in an office of Human Affairs,' Sauvigne remembers, 'brainstorming about what we were going to write on the posters for our speak-out. We were referring to it as 'sexual intimidation,' 'sexual coercion,' 'sexual exploitation on the job.' None of those names seemed quite right. We wanted something that embraced a whole range of subtle and unsubtle persistent behaviors. Somebody came up with 'harassment.' Sexual harassment! Instantly we agreed.
That's what it was. (Brownmiller, Quoted in Fricker 2007, 149-50) We might think of what Brownmiller is describing here as a version of Neurathian/Collinsian conceptual engineering. We have a community developing a concept in a safe space (the office of Human Affairs) with a particular set of desiderata (wanting to make sense of a wide range of these behaviours, wanting a name for phenomenon with which to pose a legal challenge to its perpetrators, wanting to capture the sexual nature of the behaviour and so on). Successfully developing such a concept, the feminists not only rectify a hermeneutical injustice (as Fricker 2007, famously goes on to detail) but also help to begin to pose a legal challenge to perpetrators of the behaviour, and to institute protections against such behaviour. Of course, we can easily imagine that cases similar to this may arise in the future and suggest that the Neurathian/Collinsian methodology is precisely the right one to use, as demonstrated by the case of the concept of sexual harassment.
Of course, these are not the only two cases of people undertaking a conceptual engineering project, 18 and it is not always going to be so clear cut which methodology ought to be adopted. Indeed, in some cases mixed methods may be what is wanted. However, by seeking out new and (as I have tried to do in this paper) revisiting already existing but sometimes overlooked methodologies, I take it that we can make the decision over what kind of project to undertake easier. 19