The Historicism of David Deutsch

Elliot Temple
,
21

Maybe you can help sort this out. I haven’t read Moon’s article since I wrote my article around 6 months ago. I haven’t read PoH or BoI in full for a long time. I reread some of PoH recently but a lot of the book is wordy and doesn’t seem relevant to our specific discussion.

I believe historicism encompasses multiple themes, not all of which must be present, which helps explain why you have a list of 3 versions.

Broadly speaking, Deutsch is making grand historical claims, including predicting the future course of history and making “prophecies” (predictions that you can’t make without predicting the future growth of knowledge).

Deutsch also does something similar with the past. He’s mainly ~predicting the past based on these abstract/grand theories, like he does for the future, rather than speaking about it based on historical research. His claims about the past are done mainly as armchair philosophy, not using evidence and knowledge of the past, so they have similar status to claims about the future. The past is more accessible than the future because you don’t have to predict the future growth of knowledge, but that only works if you actually have access to and use the knowledge from the past.

Deutsch presents many of his claims as logical, scientific or factually historical, and based on things like the logic of the situation or the logic of epistemology. In general this would make them non-historicist. However, this presentation is often misleading. A lot of his arguments are nowhere near rigorous, comprehensive logic. E.g. the static/dynamic meme dichotomy isn’t rigorous and IMO falls much more into the category of grand historical armchair philosophizing than logical analysis. The underlying concept (thinking about replication strategies of memes) is good, and parts of the analysis are good. But the analysis as a whole is inadequate and should have been presented as an early stages research project instead of as sweeping and very strong claims about the past and future of societies.

While Deutsch tries to present some of his claims as natural laws of memes/epistemology/creativity/etc parallel to laws of physics and based on both logic and research (to check that the historical record thoroughly matches the claims with no counter-examples), they substantively aren’t. He didn’t do much historical research and he didn’t do the kind of logical work that actually rules out alternatives, just the kind that makes a decent initial case to consider.

For some of what Deutsch missed re logical argument, see Curiosity – Third Type of Meme: Static Companion Memes

For history you might be able to get critiques at https://www.reddit.com/r/AskHistorians/ I would bet the majority of historians there would be unhappy with some of Deutsch’s history claims.

Also, without even worrying about history details, there are questions like: Was the Roman empire a static or dynamic society? Didn’t they make a decent amount of progress but also have horribly repressive, irrational parenting? It looks kind of like a mixed society. Our current society also seems mixed with varied parenting practices. But some of Deutsch’s comments don’t seem to allow for mixed societies while also not really addressing how long transitions take and how there can be a transition state when there are only two types of societies.

Specialist Creators with Small Audiences
Specialist Creators with Small Audiences
22

i’ll try.

one thing jumped out: you said claims based on the logic of the situation or the logic of epistemology would in general be non-historicist. i agree.

the BoI quote you cited reads like that to me. static societies always have traditions that disable creativity is a claim about what it takes for a society to remain static, not a prediction about where history is going. by that criterion doesn’t it come out non-historicist?

23

See my paragraph opening with the below and other paragraphs that follow up on it:

Deutsch presents many of his claims as logical, scientific or factually historical

24

seems like i misunderstood you. i reread that paragraph. let me ask rather than assume:

is the issue that the claim isn’t argued rigorously enough? if so, would it become non-historicist if it were argued more rigorously?

25

i went back to PoH. Popper distinguishes between two approaches: historicist social science, which seeks laws of historical development and transition between periods, and technological social science, which asks what institutional arrangements can achieve or prevent certain outcomes.

where does Deutsch’s static/dynamic framework fall according to you?

26

seems like i forgot to flag this: deutsch’s response to moon’s historicism critique was “Most of it is accurate.” not sure whether deutsch is agreeing with the historicism charge itself or whether moon’s structural similarity analysis gets his framework right

found another relevant part from PoH section 27: “But what about those who see that trends depend on conditions, and who try to find these conditions and to formulate them explicitly? My answer is that I have no quarrel with them.”

deutsch’s static-society claim seems conditional to me. if a society remains static, it must suppress creativity. that is more specifying conditions than prophesying where history is going, no?

27

went through PoH and BoI again and i see where you’re coming from. PoH more about identifying a syndrome than one strict criterion.

this quote jumped out:

Since the sustained, exponential growth of knowledge has unmistakable effects, we can deduce without historical research that every society on Earth before the current Western civilization has either been static or has been destroyed within a few generations.

that’s not just a conditional claim about what static societies do. it’s using an abstract theory to sort essentially all past societies without historical research. that seems much closer to the armchair problem and sweeping universal you raised. i still don’t know whether that makes it historicism rather than overconfident social theory, but it makes your worry clearer to me.

on your point about deutsch framing things as natural laws: popper’s anti-historicist sections seem especially concerned with laws of historical development and treating long-term trends as if they had natural-law universality. popper’s own open/closed society distinction also seems structurally similar in some ways, so i’m trying to understand where the relevant difference is.

on your rome point: your 2019 companion-meme post might be the missing piece. how’s this analogy: ideas can depend on other ideas like code libraries / callable APIs, so some memes can rely on a deeper criticism-suppression layer instead of implementing that whole strategy themselves. so a society could have something like a locked-down kernel/policy layer plus a richer application layer. the core static memes control what kinds of criticism or variation are allowed, while companion memes can compete for replication bandwidth in other ways inside that environment.

that also helps with rome. rome could look mixed not because it is halfway between static and dynamic, but because the application layer was thick (engineering, law) while the kernel/policy layer was still static. and your update about static memes doing within-mind evolution while controlling the selection pressure seems relevant too: maybe static societies can allow local optimization under controlled criteria without allowing criticism of the core. that seems more nuanced than the simple static/dynamic presentation.

is deutsch’s mistake mainly that the binary itself is wrong, or that he treats a rough explanatory distinction as if it were a clean phase transition? do you think the enlightenment was a genuine phase transition the way deutsch treats it, or more gradual than the static/dynamic framing allows?