Three-Toed Sloth

Books to Read While the Algae Grow in Your Fur, May 2018

Attention conservation notice: I have no taste. Also, I have no qualifications to opine on photography, American rural sociology, Aztec history, translations of medieval epics from the Italian or the Persian, or the history of science.

David Plowden, A Handful of Dust: Disappearing America

These are photographs of great beauty and melancholy, and it is sad to drive through decaying, increasingly-abandoned relics of what were once modestly prosperous communities. But I want to pick a fight with Plowden's introduction. He talks over and over about grain elevators and railroads and feed-stores, and then blames the decline of these towns on globalization and the imperatives of profit-making industry. Well, what on Earth does he think raised up these towns centered around grain elevators in the first place? They were never self-reliant, no matter what they told themselves --- "globalization enters the village through the railroad", as it were.

And still, Plowden makes it sad to see how the world has moved on from these places. §

(Thanks to Jan Johnson for the gift of this book.)

Auston Habershaw, Dead But Once

Mind candy fantasy, latest in a series where a con-artist and rogue is being operantly conditioned into something like heroism; here by an escalating series of catastrophic successes. §

Jacques Soustelle, Daily Life of the Aztecs on the Eve of the Spanish Conquest

This is a beautifully written social history of the Aztecs, with necessary glances at the political setting. While I am sure that many of the details must have been superseded by more recent research, the main outlines seem to be matched by more recent books, which continue to cite it favorably, e.g., Clendinnen. (The style is evidently the kind of thing d'Ormesson was parodying, or channeling, in The Glory of the Empire.) §

Dante Aligheri, translated by Robert Pinsky, The Inferno

Comment on my part would be both incompetent and impertinent, so I will just say I enjoyed reading this translation, and that Dante's God is a moral monster. §

George Saliba, Islamic Science and the Making of the European Renaissance

This is a fascinating little book (about 250 pages, plus extensive notes). Despite the title, the bulk of it --- the first five chapters, or ~190 pages --- is about the origins and the internal development of scientific, especially astronomical, traditions within Islamic civilization itself.

Saliba finds what he calls the "classical narrative" about those origins --- that they arose from simple contact with Byzantium and Sassanian Iran, and perhaps especially with isolated "pockets" preserving Hellenistic intellectual life into late antiquity --- inadequate. (He unfortunately never really quotes proponents of the "classical narrative".) His biggest point is that there was basically no intellectual production in the sciences in those empires, above a very elementary level, in the centuries immediately prior to the Muslim conquest, and that when the translation movement did get going, it went straight back to much earlier, and highly sophisticated, works --- Ptolemy and Galen, from the second century, to say nothing of much older ones like Aristotle, Euclid and Plato. The contemporary intellectuals of Byzantium, Saliba says, shows very little acquaintance with such authors, and so the Muslims could hardly have picked up on the classics just by osmosis.

Instead, Saliba posits a different scenario. This rests on the fact that the fiscal administration, or diwan, of the Umayyad Caliphate continued at first to be conducted in Greek (in the west) and Persian (in the east). Presumably there was a lot of continuity in personnel, too. But at some point, the administrative work (and, presumably, records) got translated into Arabic, which, considering what was involved in taxation, would have kick-started a need for mathematical terms and texts in Arabic. It also, Saliba speculates, put the fear into a lot of the old bureaucratic families, who had been relying on their inherited knowledge of how to run the bureaucracy in Greek or Persian. They, or their children, responded by digging back in to their heritage to come up with ever more impressive feats of technical knowledge, to get an edge in the now-intensified competition for bureaucratic positions. Hence the re-discover of barely-remember works. Hence, also, Saliba says, the oldest Arabic translations we still have of writers like Ptolemy are into highly fluent Arabic, by translators who are clearly very familiar not just with Greek technical terms, but also with the science involved.

Saliba then goes on to consider the development of "the science of configuration", i.e., of the configuration of the heavenly spheres, i.e., of what we might call theoretical astronomy or cosmology. Hellenistic astronomy had left a clear agenda, which was to account for the movement of the planets (the Sun, the Moon, Mercury, Venus, Mars, Jupiter and Saturn) across the sky, in terms of the rotation of spherical bodies which carried the planets around the Earth, which sat immobile at the center of the universe. (The Earth was at the center not because it was especially important, but because the element of Earth naturally tends towards the center, the point of heaviness.) The heavenly bodies move on spheres, because spheres are perfect shapes, and those spheres are to move uniformly, because in the heavens nothing either starts or ends, but rather continues eternally just as it is. All of this is laid out philosophically in writers like Plato and Aristotle, and these are the sort of principles which Ptolemy said he was following in constructing his astronomical models.

Attentive and critical readers can see that Ptolemy botched this job very badly. Leaving aside in-consistencies between his different books, the fact is that he simply wasn't able to carry this program out. In particular, he made a lot of use of "eccentrics", i.e., spheres whose center was not the center of the Earth. Worse, in a way, were the "equants", points which spheres were supposed to rotate around, sweeping out equal angles in equal times. But there is no way a circle can do this around a point other than its center, without speeding up and slowing down. Conversely, if a circle is rotating around its own center at uniform speed, it cannot sweep out equal angles in equal times around another point. On top of all this, the sky is of course two-dimensional, so Ptolemy first proposed one set of models, for the east-west motion of the planets, and then another for their north-south motion, but the latter models will screw up the former. All in all, it's a mess, which you couldn't possibly take realistically, because it's simply not self-consistent.

As Saliba makes abundantly clear, there were many Arabic-using astronomers in Muslim societies (or, as he puts it, "Arabic astronomers") who saw these problems very clearly and explicitly. There was indeed a stream of books called "doubts about Ptolemy" or "problems with Ptolemy" , part of a larger genre of books which raised scientific (not obscurantist) "doubts" about many of the classics. What is even more interesting is the tradition of "the science of the configuration", which tried to work out new planetary models, which both respected the Aristotlean constraints (spheres, no eccentrics, uniform motion and hence no equants), while being internally consistent and matching observations. Interestingly, much of this work was done after the Mongol invasions and the sack of Baghdad, under the patronage of the Mongols, the great observatory at Maragha being patronized by Hulagu Khan himself.

At a technical level, this work involved ingenious new mathematics. One prong was a device for replacing equants (at least to a high degree of approximation) with small but uniformly-rotating epicycles. The other, potentially much more significant, is a way of superimposing multiple uniform circular motions to generate uniform motion in a straight line; this originated as a way of reconciling the north-south motion of the planets without interfering with their east-west motion, but had much more far-reaching implications, since, at least implicitly, it undermined the distinction between circular motion, natural to the heavens, and rectilinear motion, natural on Earth. This device, which has come to be called the "Tusi couple" in the modern literature, can also be seen as a step towards Fourier analysis, where arbitrary trajectories are represented as combinations of uniform cycles of differing speeds and amplitudes. (Saliba does not make this last connection.) The last of the astronomers Saliba discusses also makes the very interesting step of presenting multiple models for the same planet, with a claim that they can all save the phenomena because they are all observationally equivalent.

This was sophisticated and innovative work, self-consciously going beyond what had been inherited from the Greeks. If there was any lingering thought that the scientists of the Islamic world just preserved the classical legacy, this sort of historical study should lay it to rest. (Saliba scrupulously says that he has only established this for astronomy.)

A natural and interesting question is whether this work had any impact on European science, when that emerged in the Renaissance, and in particular on the scientific revolution. In fact, many of the planetary models proposed in the "science of the configuration" also appear in Copernicus, also with the Tusi couple and the various related lemmas. This is suggestive but not, to my mind, dispositive. After all, by Copernicus's time, Latin-writing astronomers were had also thoroughly assimilated Ptolemy, Euclid and Aristotle. They thus faced the same problem, of explaining the motions in the (same!) sky using models that passed "physical" (Aristotlean philosophical) muster, with the same resources (Euclidean geometry). A certain amount of re-invention is to be expected. What does seem dispositive to me is that Copernicus's proof of the Tusi couple uses the same diagram, with the same points marked by the same letters, as Tusi's proof *. If I saw something like this from a student, I wouldn't believe it was independent work, and I don't believe it here. Saliba, properly, acknowledges that this was pointed out by Swerdlow and Neugebauer in their work on Copernicus, but he insists its implications haven't been properly assimilated, and I think that's right.

The only possibilities are (1) Copernicus had access to a source which descended from Tusi, or (2) Copernicus and Tusi both descend from the actual source. There is no evidence that Tusi didn't invent the Tusi couple, so we're left with (1). However, neither Tusi's work nor any of the other Arabic-language astronomical works descended from him are known to have been translated into any European language by the time Copernicus worked. It is possible that Arabic manuscripts of these books made their way to Latinate scholars, especially in Italy, where Copernicus spent much of his career, but there is also no evidence that Copernicus knew Arabic. There would have to have been some intermediary person who both knew enough Arabic, astronomy, and Latin to be able to pass on the contents of these books. Saliba floats a few names to indicate that such people existed, but cannot pin down to any one person who could definitely have been Copernicus's intermediary. He is also very clear that, at least so far as we currently know, nobody working in the "science of the configuration" anticipated Copernicus in heliocentrism.

Saliba's scenario for the origins of Arabic-language science doesn't convince me. He seems to be relying heavily on one old source about the change in the administration, and his arguments for why the Arab Muslims couldn't have learned from the Byzantines or the Sassanians would seem to apply just as well to the (supposed) old bureaucrat families. Even if we grant those families suddenly faced the incentives he said they did, why should they have been capable of responding in the way he imagines? None of this means that the scenario didn't happen, just that this seems very speculative and slightly internally inconsistent. About the accomplishments and contributions of the medieval Muslim astronomers, on the other hand, he seems on very firm ground. §

*: Or, to be precise, the Roman letters corresponding to Tusi's Arabic letters, with one exception which is very plausibly confusing two similarly-shaped Arabic letters. ^

Ferdowsi, Shahnameh: The Epic of the Persian Kings

Translated and adapted by Ahmad Sadri; print version illustrated by Hamid Rahmanian, audiobook read by Marc Thompson.

This is the great epic poem of Persianate culture, and one of the great books of the world. It's self-consciously the national epic of "Iran" (= the whole Persian-speaking world, not just the modern country) from the Creation down to the Arab-Muslim conquests, with special emphasis on the age-long conflict between Iran and "Turan", the country of the Turks. One of the many ironies of it is that Ferdowsi composed it a thousand years ago under the patronage of the ethnically-Turkish Sultan Mahmud of Ghazni, and in doing so helped establish modern Persian as a literary language. It's unreliable as history --- among other things, it anachronistically projects the mounted, armored knight, who was indeed a Persianate invention (stirrups FTW) back for millennia, and the account of Alexander (Iskandar) is very tangled (making him and Darius [Dara] half-brothers!) --- but that's hardly the point. Rather, it captures an ethos, and a wealth of mythology, in beautiful language, and both codified and helped shape self-understandings for a millennium.

This work in English is a translation and condensation of roughly the first 2/3 of the epic, down to the part where it transitions from being almost entirely legendary to quasi-historical. (Specifically, it ends with the death of the great hero Rustam.) The translation is into very readable contemporary prose, with occasional bits of rhymed verse to mark the places where Ferdowsi allowed himself explicit morals or self-depictions. The print version is lavishly illustrated, with new art mostly following the tradition of Persianate miniatures. The audiobook is well-read, with, crucially, good pronounciation*, but includes quite a bit of music and sound-effects, aptly described in the introduction by Francis Ford Coppola (!) as a bit over the top. The result is one of the two best English version for general readers I've run across. §

(With thanks to Jan Johnson for the printed version.)

*: I was a bit thrown by hearing all these names pronounced in the Farsi rather than the Dari manner, having grown up hearing "Shahnama" or "Tahmina" instead of "Shahnameh" or "Tahmineh". But I daresay this will not bother most listeners.

Books to Read While the Algae Grow in Your Fur; The Commonwealth of Letters; Islam and Islamic Civilization; Afghanistan and Central Asia; Physics; Writing for Antiquity; The Beloved Republic; Scientifiction and Fantastica