Synthetic philosophy, restated

Eric Schliesser, University of Amsterdam (Political Science) 💬

Synthetic Philosophy

• Reinterpretation of Republic's themes, in a way secondary literature hasn't assimilated

• Intellectual (hyper)specialisation a.k.a. The Advanced Division of Cognitive Labour

• 2019 paper went viral but was misinterpreted

• Schliesser (2019) Synthetic philosophy. Biology & Philosophy.

• Philosophy of science are oddballs in the profession (high status-ish), clear divide and other side aren't reading each other

• Some special sciences are neglected (e.g. political science/chemistry)

• Extensive division of labour creates important social problems/externalities

• Especially when it comes to uptake of specialist knowledge in policy or "world-making" (when people don't understand or appreciate expertise)

1st motive: philosophy of philosophy

• Meta-philosophy: "The Philosophy of Philosophy", "Philosophical Progress"

• Creating necessary and sufficient conditions for universal generalisations

• Modelling conceptual/modal space (LA Paul/Williamson)

• Focused on arguments and progress through discovering what option to drop in a dilemma/trilemma (cf. Stoljar)

• To other philosophers, philosophy of science often looks like bad conceptual analysis

2nd motive: extensive division of cognitive labor

(via The Great Endarkenment, see his blog)

• linguistic and cognitive barriers as we hyperspecialise

• Has an effect: most of our beliefs are authoritated by other experts (based on authoritative testimony of others). Epistemologists are nervous about testimony

• If we don't understand what other scientists are talking about we get vulnerable to things like the replication crisis (not only reason: incentives matter too)

• Creates policy problems for administrators, grant agencies, government policy, journalism. Especially apparent during the pandemic.

• Written on this: "Novel externalities" (2023)

• It explains why university leadership increasingly focus on procedure (they can't feel confident about substantive decisions)

• Even great science journalists make blunders when writing about your subfield

• Administrative law (and judges) farm this out on a hope they chose the right person/metric/procedure

Synthetic philosophy as a style of philosophy

...aiming to bring together insights, knowledge and arguments from the special sciences aiming to offer a coherent account of complex systems and connect them to a wider culture or other philosophical projects (2019)

"Integrative conception": "going hermeneutic on oneself"

• Integrated scientific image of different specialisations

• However... the paper also said that synthetic philosophy requires a general theory (e.g. Darwinism or game theory or info theory, perhaps Bayes' theorem) thin and flexible enough to be applied in different special sciences but rich enough that when applied it allows for connections to be developed among them (pp. 7-8)

• Argued Darwinian philosophers (e.g. Dennett) have been doing this

• Thomas Pradeu had argued similarly in 2019: "Why science needs philosophy"

Anticipated by Lewens

Partially anticipated by Tim Lewens (2014) reviewing Kim Sterelny (2012) The Evolved Apprentice: How Evolution Made Humans Unique in The British Journal for the Philosophy of Science

• That general theory is a particular kind of expertise

Plato's Republic

Plato's Republic: division of cognitive labour is foundational to the Republic, the cause of all political life (healthy city), the effect of human diversity (which is also cognitive in character)

• In the "ship of state" parable of the Republic, democratic elites are attacked for claiming that there is no skill that can be taught to pilot the ship of state and the people with real skills don't have them, and they themselves the people they like do have it

• Crucially the owner of the ship (the public, the demos) cannot recognise real skill even when face to face with it (tragedy diagnosed by Socrates)

• This comes back in the Statesman dialogue, occurring over and over again in polemics with sophists. A non-trivial obsession in Plato's writings.

Socrates and the poets

Reminder of book X: Socrates returns to call for censorship of the poets (who corrupt the young)

• Poets do not know what they describe (602b)

• That's important as poets are major producers of culture in public life

• Socrates claims this ignorance sowing stems from lack of distinction of knowledge, its lack, and imitation

• This generalises to all speakers: not just the poets

• In our day, this is a long list of people with degrees: journos, lawyers, consultants, pundits, internet gurus etc

• In broader context, Socrates seems to assume that regular users/consumers of a particular craft (e.g. users of sofas) are eventually able to make judgements on quality, but believes they can be fooled

• Among the topics of poets' ignorance are "the greatest and most beautiful things... warfare, generalship, city government [or city administration, public policy]"

A surprising textual detail: Damon

Damon is introduced as a technical expert on the representation qualities of music or rhythms and its role in character development

• Socrates says we will appeal to Damon a) to save time and b) he has knowledge Socrates lacks

• Damon's expert terminology is a barrier to understanding: you have to know it to judge his expertise

• Socrates' skill in dialectics is a way to determine real expert from non-real

• Unclear if Socrates is being sarcastic (according to some interpreters)

• A book later, Socrates returns to Damon and makes a striking claim: a change to a new type of music is potentially destabilising society. "For the modes of music are never disturbed without unsettling of the most fundamental political and social conventions"

• i.e. if you shift the culture you risk politically destabilising society

• He says "as Damon affirms and as I am convinced" (no sarcasm)

• Damon's authority as accepted expert is accepted here

• Musicology, music theory is a very old science. Damon's musical expertise is central to the art of government

Plutarch and Damon

Plutarch also interested in Damon (On Pericles)

• Ancient interest in Damon, classical scholars and musicologists have an interest

2 models of how to think about division of cognitive labour

Kallipolis and nth best model

• Best model (kallipolis), reorganises society, breed elite knowledge class (in which all sciences are philosophical), long training, esprit de corps, no property, censorship prevents distortions produced by political life and culture (to keep epistemic environment sanitised), philosopher-kings

• Nth best model, where expertises are overlapping and have to defer on occasion (anticipates Polanyi). Socrates' skill [in dialectics] facilitates this

There are other theoretical models

• Leibniz (Max Planck Institute): government should organise research institutes

• Bacon's Royal Society project

• Encyclopedism (HG Wells, Diderot)

• etc

There is a challenge of securing the fruits of the advanced division of cognitive labour, for any society. There is always a highly politicised agent incentivised to prevent you having those fruits.

Plato may be the source of the ur-myth that all sciences are philosophical in character (i.e. that philosophy birthed the sciences)

Synthetic philosophy is a kind of umbrella term: "style/kind of philosophy that presupposes or develops expertise in a general theory thin and flexible enoguh to be applied in/to different special sciences but rich enough that when applied it allows for connections to be developed among them with the aim to offer a coherent account of complex systems and connect these to a wider culture, the sciences or other philosophical projects (or both)

It can be taught to others without mystery

A list of things that can be this synthetic glue (given above, game theory, info theory, Bayes' theorem, causal modelling, agent-based modelling etc) also category theory and homotopic type theory.

Creates a class of philosophers which can communicate with stakeholders.

Commonality of different styles not in itself interesting, necessarily.

Originally (2019) conceived as an alternative to analytic philosophy, but compatible with it.

Given the non-unity of contemporary science, synthetic philosophers may well disagree with each othernot just on methodology on substantive metaphysical or ontological issues.

Another worry: is there a more precise way to characterise a synthetic theory?

Q&A

Is the task of synthetic philosophers to develop a common vernacular?

Q) Is this saying that the task of synthetic philosophers is to develop a common vernacular?

A) I think that's a useful way to think about it, but the vernacular is itself more technical and more specialised than we come to think of when we think of pidgins and common vernaculars (lingua francas). The nice thing about lingua francas is that the barriers to entry is relatively low. With pidgins it grows up more organically [than a common language like English] than in "trading ones", it develops between trading users. The synthetic philosopher is analogous to that but distinct, as the barriers to entry is going to be somewhat more specialised (not necessarily hyperspecialised, if it takes a PhD to understand game theory then we lose the benefit), but more so than those in the analogy. On a sliding scale the analogy is useful, but "close but not quiet".

Q) That's helpful. At the very least it has to enter into the theory otherwise you'd just learn that instead.

A) In actual fact, the expertise of the synthetic philosopher is about bits of science or modelling that are themselves super specialised. In an ideal world they could be contributing and not merely consumers of it. We do run the risk that the synthetic philosophy is so esoteric that the problem starts becoming like a fractal. Mathematicians and computer scientists keep promising us that they'll solve this in a general way, but it hasn't happened yet.

Q) Interesting whether you pass through being a synthetic philosopher and pass through into hyperspecialisation... Maybe they can be both...

A) I see this happening with friends, sometimes they "come back from the brink". We don't want to lose the "freedom to roam". To become hyperspecialised for a while, but then a tug back, "now explain to the rest of us".

Are mathematicians synthetic philosophers?

Q) You used a number of meta-theoretic terms when outlining this, but you also get these meta-theoretic terms occurring in mathematical practice, mathematicians who are utterly innocent of philosophy. What sort of claims are you making here?

A) I hope to be entertaining. I get off the boat at your second sentence. Mathematics reguarly goes through phases where it reflects on themselves. Not all mathematicians go down this route, but in my view (a Kuhnian idea), philosophy can show up in lots of places that are not professional philosophy, and I wouldn't want to say that the mathematicians who have reflected on their practice... what they're doing is not the real deal. It's possible that the terminology I'm using is mathematical, but in the history of mathematics they borrowed terminology from philosophy.

What are the 'bits' held by the 'integrative glue'?

Q) 2 parts are the bits and the glue, I want to ask about the bits. What counts as a special science? Is any sort of domain with distinctive expertise, hard to understand from the outside, or is it important for it to be a proper science?

A) This is a terribly important question I don't have a good answer to. I might end up deferring to the specialists who have spent their time thinking about the demarcation problem. Do you have a suggestion?

Q) There are lots of specialisms, literary critics for instance.

A) Synthetic philosophers are an expert on particular method that potentially have a home in more than one discipline. I do think that in the humanities, there are particular methods used both through history and now, that are of a sort very much like I have in mind that are the domain of synthetic philosophy. (Hans Butthe, ...) there is genuine technical expertise that can both be written out, taught to others, and has been for hundreds of years. Whether there are intellectual pursuits in other ways... I can't say. Specialist knowledeg passed on for generations can be passed on for generations, whether that makes it a science... There are disciplines where it's questionable. There's a sense in which I want to screen off what the language is about from what the expertise is.

Q) Right, so it's primarily about the glue.

A) Right, but no mystery is allowed, no "and then something special happens". I'm not against mystery but that's not what the synthetic philosophy thing is about.

What is the relationship between 'glues'?

Q) I wanted to ask about the relationship between glues as it were. It seems as though each synthetic philosophy will bring together a certain set of special sciences and confer on them a certain... status? Intelligibility? (So that's the question). But then there are different glues which bring together different sets of sciences... So are we to think of these as different possibilities, or is there as it were something special about them...?

A) It's interesting that you put it in terms of intelligibility or... [Ed.: status]. But in my own mind I'm particularly interested in coherence constraints, but in terms of the glue a certain kind of knowledge or skill why I used Plato] is important... One ought to be able to do things with the material and do it reliably, and get others to see what one is doing, so that certainly involves a kind of understanding. And the 2nd question, I think the answer is certainly yes: the list of things they are gluing is very distinct. Daniel Stoljar commented on a draft and said haha Eric, synthetic philosophers have very different commitments. But I don't see it as a problem. That doesn't mean that one might well want a classification of the glues. That becomes a mathematical problem: how much ontology/generality/fruitfulness is smuggled into the glue? I don't have this Aristotelian/Kantian classificatory gene that some people have. I can imagine a whole listings of what glues are and hierarchy among them, what I very much doubt that a single synthetic philosophy hierarchy of glues is possible. If one arose in 500 years I would be very surprised, and dismayed.

How are we to train synthetic philosophers?

Q) I am curious in how we are to train people in synthetic philosophy. Should we train people as philosopher first, to cut their teeth, or another way? Both?

A) This is a cool question in philosophy of education, of knowledge, and allows me to dream Dewey style, Plato style cult of education. In practice what we have seen is both. First you have a solid founding in philosophy, then do a PhD in X, but in practice the last 100 years of analytic philosophy has been a bit of a sponge both in methods and in people coming in. You can sneak into analytic philosophy in many ways. I think to your dean you can say look a department that specialises in synthetic philosophy can talk to different fields in intelligible ways, and work at the intersection (deans love this). Part of what drove this was thinking about the fact that a lot of people I admire ended up in these quasi-interdisciplinary centres, but were worried about their graduate students (that's part of the real world). Given the state of higher education, whatever works locally you should use. But I do think that for all such programs is I think what's bad if a dean says here is what should work... as this has an opportunity cost (people unable to talk to each other). If synthetic philosophy has fecundity in generality it's also the case that getting up to speed in it doesn't take 6 or 7 years to get going.

Why is this a specific philosophy?

Q) My question may sound like a provocation but is a question. Why is this a specific philosophy? I am in a government department [at LSE] it is very common to use a 'glue' like game theory to explain economy. These people will not say they are philosophers, they will say they are economists for the clout. Are these people just annoying synthetic philosophers?

A) The way I've been positing synthetic philosophy is both the expertise and the reflection on the tool itself. Von Neumann, great mathematician, but also had philosophical work... Case by case, are they just applying the tool? Is there an interest in seeing if bits of sociology and economics can be melded together? [Downs?] Using economic tools in political science, almost informally, thereby made things visible which people in [both sides] which people weren't paying attention to. There are a number of cases where you can come up with people doing something analogous to synthetic philosophy. It doesn't have to be methodological, it can also be a development of the tool into another area, or explaining the tool to people. Given my interest in how do we actually certify this expertise to the expert world, we've now bureaucratised that, and we're approaching a dead end so have to think about new ways of doing this.

Why pursue analytic conditions for synthetic philosophy?

Q) I wanted to ask about the apparent rapprochement between synthetic and analytic philosophy are. I was thinking why would you want to hear the necessary and sufficient conditions for synthetic philosophy? Surely the idea that there would be a static concept to get necessary and sufficient conditions [given the diversity of glues] would be the kind of shibboleth you'd want to abandon.

A) Maybe you're right, but I think it'd be a little awkward if we all pretended to understand what synthetic philosophy is, but were in fact talking past each other. I am interested in Amy's question of 'what are the different glues?'. In climate science you really want to understand the relationships between different kinds of models, there in those fields it's very interesting to understand the different glues, keep track of them and their hierarchies. While I agree a nec. and suff. condition is true, having more than an heuristic sense of them is important. In the contemporary Bayesian epistemology, which scoring method you use to score your tool turns out to be very important. Then you want to ask what are the effects of different scoring methods in different contexts. That's what I think synthetic philosophers need to understand, almost a cost benefit analysis of difference glues. If you go around philosophy land, there are a lot of people excited by a formal apparatus. They treat it as a hammer for every nail. They don't explain to the audience in which contexts their hammer has particular (dangerous/unknown) consequences. When I ask about the duty of synthetic philosophers this is what I think about. It isn't nec./suff. but maybe the inductive risk of a particular apparatus. You have to model that and have a good enough understanding of what the apparatus is. This will be a list of characteristics that are indicative of this type of glue and not another. I think this is an important project to pursue.

Do synthetic philosophers have a role as public philosophers?

Q) I'm wondering what your view is on what the role of public philosophers should be, and if synthetic philosophers have any role in communicating back to the public. This relates to Michael's question of how we train in philosophy.

A) I don't see myself as a public philosopher. I do think the terminology of public philosophy is very open. They really think of it as a way of communicating with a wider public. In our contemporary discourse, it's genuinely if it gets [media coverage, NYT/Guardian]. This is fine but my motive to say "hey there is such a thing as SP and it's about "glue expertise", I think one potential payoff of thinking about this in a serious way is those people can play a role mediating expertise between other experts and with a wider public. Whether grant agencies, administrators, a learned audience, or beyond. I think those are very different projects. Explaining cost benefit to a group of public servants is very different from writing a book that will hit the bestseller list. In both cases it can be that nobody is talking about the glue, other times they are. Hence I focused on the Republic. The question of how knowledge relates, how expertise relates, is one of the questions on my mind, and definitely where SP is an answer, but not the only answer.

What is the role of synthetic philosophy in democracy?

Q) Similarly to the previous question. You started off with a long political discussion, I wonder if you have thoughts on the very prevalent epistemic discussions, where we have to vote on what people are doing, and whether synthetic philosophy can have a role in what we are doing.

A) Someone like Neil Levy really does think that we can have epistemic environments, social spheres that are relatively robustly truth-tracking. One reason I'm attracted to Plato's story is the branch of liberalism I find myself in is sceptical of the possibility of truth tracking of the public being reliable. Not nihilistic, but realistic. This is common prior to Rawls. Contemporary liberalism, Habermasians, are very optimistic about how much truth public opinion can produce. That doesn't mean we have a solution, but insofar as our society has a lot of intermediary institutions between both individual voters and experts, where there's a chance to impact it. Bruno Latour: when people are not paying attention, a lot of ways expertise can be brought to bear. When things go wrong we start paying attention again. I'm not a defeatist about this. The way you've formulated the question, I think this is helpful. I think voting can be expressive (and protesting), not about epistemically truth tracking necessarily or even primarily. If you do think it's constructive to think about voting like that, good for you. A lot of my colleagues do too.