Here is a text I might have included in the book, if I had known about it in time:
Michael A. Coxhead, A close examination of the pseudo-Aristotelian Mechanical Problems: The homology between mechanics and poetry as techne, Studies in History and Philosophy of Science 43 (2012) 300-306.
Mathematics without Apologies cites the pseudo-Aristotelian Problemata Mechanica in a footnote to the discussion in Chapter 10 of the historical distinction between art and science:
When we have to produce an effect contrary to nature, we are at a loss, because of the difficulty, and require skill [techne, art]. Therefore we call that part of techne which assists such difficulties, a device [mekhane].
The word mekhane had already been examined in Chapter 8 as one possible source of the modern propensity to classify some ways of doing mathematics as “tricks”, and there it served as an intermediate term between techne (art) and episteme (science). But (as befits a trickster) it has multiple identities. Pseudo-Aristotle classes mekhane as a “part” of techne. Coxhead sees this as analogous to (genuine) Aristotle’s view of poetry as techne, more specifically poietike techne. For our purposes, the most striking sentence in Coxhead’s article is this one:
Aristotle conceived of both mathematics and poetry as offering universal knowledge.
I don’t know Greek and can’t read the ancient texts, though with a dictionary and a few articles like Coxhead’s handy I can piece together a few words at a time. So I understand, for instance, how the word “poetry” relates to the Greek root poiein [to make, to produce]. Confusion sets in when Coxhead alludes to pseudo-Aristotle’s “opinion of the human use of art (techne) fits firmly within the Aristotelian notion of poetry as a productive science [poietike episteme],” because elsewhere (as documented in my book) Aristotle draws clear distinctions between art and science.
Fortunately, philology doesn’t let itself get discouraged by a few inconsistencies, so I won’t be discouraged either.