A review of this book should start, I think, with annotations on its title. "Cosmos" refers to the universe as an ordered, harmonious whole--the opposite of "chaos." The pair "cosmos and community" ("community" being contrasted with "solitude") has three aspects: a political, an epistemic and an existential one. First, communal order--the power structure of a given society--can be legitimized as the supposed reflection of a "natural" cosmic order. Second, information about the presumed order of the universe is shared within "communities of knowledge" which coalesce into distinct, usually privileged social groups. Last but not least, projecting orderly concepts onto the inanimate world helps humans feel "at home," "in communion" with it. The disquieting vastness of the starry sky, for example, is somewhat tamed when constellations are reduced to a set of familiar figures. Perhaps because they correspond to a constant human need, such figures have proved historically durable: a present-day reader of magazine horoscopes can easily recognize depictions of the zodiac in Ptolemaic or Roman sculpture.
Anderson discusses examples of zodiacal imagery, along with some world maps and cosmological diagrams, from the seventh to eleventh centuries. He has selected works whose spatial setting or social context (patronage) is known. His book aims to show how three emergent imperial powers--Byzantium, the Caliphate, the Carolingian/Ottonian state--made use of cosmological imagery inherited from their common predecessor, the Roman Empire. The general conclusion is that Franks and Arabs actively employed such imagery in order to forge unity among disparate political elites, while for the Greeks it became a marker of exclusive--and thus potentially divisive--knowledge limited to the emperor and his court.
The book's single weakness, for which its author cannot really be blamed, is that most of the source material is sparse and correspondingly easy to over-interpret. Upon describing a monumental geographic map, one Latin panegyrist expressed "delight to see a picture of the world, since we see nothing in it which is not ours" (1): from this Anderson infers that a depiction of the starry sky upon a ceiling would have filled Roman viewers with comparable pride (2-3). Seeing buildings and trees in the mosaic frieze of the Great Mosque at Damascus must have likewise gladdened Muslim visitors, for an Arabic poet (whose exact words are not quoted) compared this frieze to the entire land of Syria--which is pretty much the same, we are assured, as "an image of the world in which nothing was depicted that did not belong to them" (33). In the bathhouse attached to his reception hall at Qusayr 'Amra, an Umayyad prince might have entertained local dignitaries; constellation figures painted in fresco upon the (steam-filled) caldarium's dome "will have served a reminder that the outcomes of their discussions had global consequences" (69). The embroidered Latin inscription Populus qui conspicit omnis arte laboratu(m) is taken to evoke "an image of the cosmos that constitutes a people"; with its figures and text, therefore, the 83-cm wide "cloth conjures an omnis populus, a people united in the act of viewing" (74-77). The miniatures in two otherwise very similar Greek manuscripts illustrate Genesis 1:24-25 with a round and, respectively, rectangular image of the earth, in both cases surrounded by water; for Anderson, the two "envision the biblical account of creation through two opposed cosmologies" (136-139). A great deal is made out of the Latin idiom ad astra (41) and of what I assume must be its Arabic equivalent (69).
The book's best-documented and to my mind finest sections are those dealing with the Carolingian period. Three sources describe a silver dining table with "a representation in careful and tiny drawings of the whole world in three connected circles"; this precious object, long since lost, is shown to have reflected the power aspirations--both intellectual and imperial--of its successive owners: Charlemagne, Louis the Pious and Lothar (35, 55-63). Four groups of illustrated Latin manuscripts all date from circa 800-820; for many of them, the exact places of production and/or patrons are known. Anderson explains how the relatively uniform astronomical imagery which these books contain propagated the same cosmological concepts "across a broad geographic expense and among members of various overlapping elites," mediating "a relationship between region and empire" (78-105).
The author astutely switches between historiographical critique (9-16, 46-48), historical narrative (21, 27-28, 97), functional analysis of buildings (29-32, 68-69), formal analysis of images (67, 102, 118, 121-23), and contextual analysis of inscriptions or longer texts (21-26, 34-35, 41-42, 70, 74, 104, 143). When dealing with primary sources, he puts to good use his knowledge of Arabic (154n.19, 158n.83, 159n.90-92.101-103.110, 168n.105), Armenian (154n.20), Hebrew (151n.30), Greek (154n.21-24, 177n.48, 180n.104) and Latin (152n.52, 160n.117, 162n.146, 164n.34.36, 165n.50, 166n.73.80). He has studied with admirable thoroughness vast scholarly literature pertaining to three separate fields: Islam, Byzantium, Latin Christendom. The only serious bibliographic omission that I spotted is Bruce Eastwood's Ordering the Heavens: Roman Astronomy and Cosmology in the Carolingian Renaissance; other studies by Eastwood are duly cited. A couple of small errors--there was no "Antigonid court in Athens" (86); the word zygos in an astronomical context means "Libra", not "yoke" (107-108)--do not affect the argument.
What useful purposes is this book going to serve? It will certainly advance its author's academic career. It would delight any reader with its splendid erudition, fine writing style, abundant color illustrations and truly beautiful page design (the work of Leslie Fitch). And it will inspire students to overstep the customary boundaries between eastern and western, Christian and Islamic art.