In the field of printing, the prophetic types of Moses and Ezekiel gave way to a different fixed form, the wood and metal block that held the form of a letter.
Typology entered natural history in the 1830s, when the famous naturalist Georges Cuvier adopted the notion of the type from the printing press to describe his new classification system, which would later be used as an adaptation of the Linnaean system.47 Cuvier’s types, unlike printers’ blocks, were meant to arrange nature in a definite hierarchy in which lesser animals were represented by species that served as examplars (or types) of their particular genre. This was a typology of synecdoche, in which a strategically chosen part had the power to represent the greater whole. The many varieties of oblong fish who had hard scales, spiny gills, and toothy jaws were best represented by the perch, their type; while the leaf-cutter ant Formica cephalotes of South America, famous for its large colonies and social behavior, proudly stood as the type for the entire ant genus.48 Historian Edward Eigen suggests that it was with rather coy self-awareness that Cuvier used different print types to show conceptual types, deploying various typefaces in Le Règne Animal to rank the categories in his prose from more to less general ideas.49
Over half a century later, when Cuvier’s types gave way to type specimens, naturalists retained their mania for connecting words, bodies, and categories by synecdoche. They made up new words, like paratype, prototype, and holotype, for different kinds of specimens. One of the first articles in The Auk, the organ of the American Ornithologists’ Union, suggested a long list of new words to be used in species nomenclature that included neologisms like “chironym” (for species names that had not yet been published) and “onymizer” (the person who successfully finds and names a species).50 Several years later, paleontologist Charles Schuchert even suggested that biologists adopt words like “protolog” for the first written description of a specimen and “protograph” for the first known picture of a specimen.51
The bodies that these words represented reflected the tension between ideals and reality that was embedded in the notion of the type. Like Cuvier’s types, specimens were supposed to represent an entire category of bodies. It was a heavy burden to place on the preserved remains of plants and animals, and naturalists carefully culled through their options to cultivate specimens that, they felt, met the ideal of a species. The modern museum preparator was expected to be a naturalist, sculptor, and artist, crafting what biology editor Frank Thone rhapsodically called “the exact form of the animal’s lithe grace, its smooth waves of muscle, every vibrant detail that existed under its skin while it was still breathing and moving.”52 The story of the process through which such specimens were selected for display in the American Museum of Natural History’s Hall of African Mammals, as told by Donna Haraway, shows just how important this interplay was between taxonomic practice and idealizations of the body.53 Haraway’s story demonstrates that the type specimen of a species was almost always male, and that display specimens were carefully curated to represent idealized postures and attributes ascribed to a given species.
This process of preparing and ordering specimens was just as important for those who collected and curated sonic bodies in the early twentieth century. But what was a musical “type”? Unlike an animal body, a song changed every time it was sung, and the recordings and transcriptions that served as sonic specimens were records of transient, singular experiences that had fundamental differences from the biological type specimen. A musical type was revealed by a whole collection of specimen melodies, not a single example—there was rarely a musical “type specimen.” Instead, the cowboy “type” was found in the melodic pattern of American ballads for one scholar, while the English “type” was discovered by another in simple uses of church modes.54 Naturalists, not to be outdone, compared hundreds of birdsongs to determine species types in the first half of the century. Aretas Saunders identified seven types in the song of the field sparrow (Spizella pusilla), and five for the song sparrow (Melospiza melodia).55 Ornithologists at Cornell University managed to outfit a truck with film recording equipment in the 1930s, translating the songs of wild birds into the visible lines scrawled across the soundtrack, images they then compared under a microscope in their own search for avian song types.56 The collector who financed that expedition, Albert Brand, sent some of the resulting recordings to a music ethnographer named Laura Craytor Boulton (one of the protagonists of chapter 3) for inspection.57 Ethnographers working with the Bureau of American Ethnology on Native Americans worked especially hard to identify song types.58 Before 1918, Frances Densmore plotted her collection of six hundred Sioux songs in tables and charts to determine five song types whose contours exemplified the Sioux tradition, while her colleague Helen Roberts borrowed Densmore’s method to identify a type for the songs used in the Pawnee Skull Bundle Ceremony.59
Like biological types, song types navigated the disparity between ideals and thorny reality. Ideals in the case of music were closely tied to preexisting racial and national typologies. Music specialists often took for granted the “decisive leaning towards rhythm” of the French, the “innocent minds” of Italian melodists, and the “strenuous thought” of German musicians, hearing their own expectations in the songs of various nations.60 British collectors Cecil Sharp and Charles Marson explained early in the century how a sufficiently large collection of songs could be used to scientifically demonstrate these associations, and eventually prove that “the scheme of a tune … develops racial traits [in time], just as the country has developed racial traits in the music of speech.”61 By the 1920s and 1930s, such empirical aspirations had reached into the laboratory, with experimental psychologists such as Carl Seashore suggesting that quantitative measurements of sonic specimens would “take us into the field of genetic studies of inheritance … and studies of racial types and the evolution of music in primitive peoples.”62
Song typologies thus occupied the center of a three-part process in which songs were institutionalized and analyzed as objects of scientific research. First, collectors gathered them in the field. Collections were then assimilated into institutions and organized through typological practices. In the final and aspirational step of this process, song types were confirmed and analyzed in laboratory settings like Seashore’s. Just as the lives of collectors shaped the songs they collected, the central step in this process of institutionalized classifications of music was also shaped by life’s ambiguities.
The gap between real and ideal in the search for song types was particularly affected by problems of notation. Many of the specialists invested in song typing expressed concern that Westernized notations were making it difficult or impossible to study songs properly.63 But those worries were often simultaneously framed by a desire that alternatives to Western scores reflect not a song’s particular details, but its ideal typological form. As one collector of Russian folk songs explained, scholars needed to reduce songs “to their simplest form and, by omitting melodic ornamentation dependent upon the individual taste of the leader, preserve only the foundation of the melody.”64 It is worth reading closely the words of Helen Roberts, as she described the problems facing sonic typology in 1922 amid her research at the Bureau of American Ethnology:
Not only is it difficult to seize upon and designate the peculiarities which distinguish certain types, upon hearing the song, but in reading over the notation they are not all equally clear to a musician, who must needs reduce the music to some simple formulae covering the structure, etc., in order to have them clearly in mind. Such a reduction is even more necessary for those whose unfamiliarity with the symbolism of printed music renders the subject still more complex. For the sake of obtaining the bald outline of the tune and the design which it formed, structurally, it seemed best for the time being in analyzing different songs to eliminate key signatures, musical notes, with their different values, all pitches less than whole step intervals, all measure bars and accents, all expression marks, in fact everything that might be considered to belong to the realm of color in music.65
Roberts suggests that although the collected
