The Poverty of Historicism by Karl Popper

I was introduced to Popper by David Deutsch in The Beginning of Infinity, and also somewhat through parenthetical references made by Nassim Taleb in his books. Both present Popper's ideas clearer than the man himself, in The Poverty of Historicism.

Deutsch quotes Popper incessantly, and the ideas he (Deutsch) talks about are awesome. Two that struck me:

  1. Most progress is incremental, and more about error correction than consistent correctness.
  2. Explanations of sociological facts (like why a war happened) are best described with other sociological facts, and not with physics or mysticism.

An example from Deutsch, on the latter: the best explanation for the presence of a copper atom in the nose of a statue of Winston Churchill will not be a physics equation describing the motion of that atom or a chemistry equation describing the form of the statue, but instead it'll be a reading of the history of actions taken by Churchill and of how those actions were perceived by those with the means to make and place statues.

When I heard that these ideas were elaborations and refinements of Popper's, I was excited to read the original. Instead, I'm now all the more grateful that word processors exist. This book reads as if the author wrote it in a single top to bottom rant, with no means for edits or revisions. And perhaps he did.

Concrete criticisms:

  1. Everything is discussed in abstractions and schools of thought: of wholes, holistic processes, Utopias, non-naturalism, Marxism, Gestalt psychology, eras, States of Society. The only time the author resolves his abstractions into concrete examples is parenthetically. And so the reader is left to try and imagine to what, in the actual physical world, Popper's ideas correspond. This is made harder now that it's ~65 years after the book's original publication.
  2. So many of Popper's sentences distract from their main expression. They have these nested parentheticals and digressions and enumerations of near identical nouns and adjectives.

    As in, rather than saying something like, "History is hard to describe in causal detail, but it's not impossible. " the author might say something like "History, in the sense of  the ledger of events that we've recorded to have had happen, is difficult, some say impossible others say a job left to the prophets, to describe with any generality, whereby describe I mean elaborate on the abstract effecting means and by generally I mean with any amount of reproducibility of terms."
  3. The author assumes his audience is a computer compiler. At one point in the book he defines the word "whole" with two competing definitions, and then labels those two definitions (a) and (b). He then goes on to make a multi-page argument about these two definitions without ever reiterating or clarifying which definition he's talking about, instead inserting back-references to those single letter labels: (a) and (b). It's then up to the reader/compiler to accurately replace each label with the correct definition.  

    For instance, he says, "Nobody doubts, of course, that wholes in sense (b) can be moulded or controlled or even created as opposed to wholes in sense (a)."
    (Note also that unnecessary and distracting enumeration of verbs, again.)

    This would be a fine statement if it were made in close proximity to the original definition, but if one makes increasingly abstract and disparate arguments about these points, and elaborates on each point for paragraphs, then several pages later the reader is left saying, "Wait, which definition was (a)?" or "Was that (a) or (b) he was talking about there?"

    Given the quote above, I just wish it were written like this instead, "Nobody doubts that the the special properties of a whole, those which make it more than the sum of its parts, can be controlled. For instance, an engine can be throttled without much understanding of its pistons. Further, nobody doubts that wholes which are merely a heap of constitutes are harder to control since they have no special properties to control. Tell me where the 50th rock might land, after 49 previous rocks have been thrown into a pile."

    Even trying to clarify that sentence myself, I'm left wondering about ambiguities in the original form that allow the sentence to be read in contradictory ways.

None of his says anything of his arguments, which I'm still trying to excavate.