Trees; evolutionary theory. They come last, in this series of essays, but were really the beginning, as my Marxist formation, influenced by DellaVolpe and his school, entailed a great respect (in principle, at least) for the methods of the natural sciences.footnote1 So, at some point I began to study evolutionary theory, and eventually realized that it opened a unique perspective on that key issue of literary study which is the interplay between history and form. Theories of form are usually blind to history, and historical work blind to form; but in evolution, morphology and history are really the two sides of the same coin. Or perhaps, one should say, they are the two dimensions of the same tree.

Figure 1 (below) reproduces the only tree—‘an odd looking affair, but indispensable’, as Darwin writes to his publisher in the spring of 1859footnote2—in The Origin of Species; it appears in the fourth chapter, ‘Natural selection’ (which in later editions becomes ‘Natural selection; or, the survival of the fittest’), in the section on ‘Divergence of character’. But when the image is first introduced, Darwin does not call it a ‘tree’:footnote3

Now let us see how this principle of great benefit being derived from divergence of character, combined with the principles of natural selection and of extinction, will tend to act. The accompanying diagram will aid us in understanding this rather perplexing subject . . .footnote4

Dendrogram from Charles Darwin's 'on the origin of species', indicating divergence of character, depicting two figures that start at point A, and point I, respectively. Note reads: let A be a common, widely diffused, and varying species, belonging to a genus large in its own country. The little fan of diverging dotted lines of unequal lengths proceeding from A may represent its varying offspring... Only those variations which are in some way profitable will be preserved or naturally selected. And here is the importance of the principle of benefit being derived from divergence of character comes in, for this will generally lead to the most different or divergent variations being preserved and accumulated by natural selection.

A diagram. After the diachronic diagrams of the first article, and the spatial ones of the second, trees are a way of constructing morphological diagrams, with form and history as the two variables of the analysis: the vertical axis of figure 1 charting the regular passage of time (every interval, writes Darwin, ‘one thousand generations’), and the horizontal axis following the formal diversification (‘the little fans of diverging dotted lines’) that would eventually lead to ‘well-marked varieties’, or to entirely new species.

The horizontal axis follows formal diversification . . . But Darwin’s words are stronger: he speaks of ‘this rather perplexing subject’—elsewhere, ‘perplexing & unintelligible’footnote5—whereby forms don’t just ‘change’, but change by always diverging from each other (remember, we are in the section on ‘Divergence of Character’).footnote6 Whether as a result of historical accidents, then, or under the action of a specific ‘principle’,footnote7 the reality of divergence pervades the history of life, defining its morphospace—its space-of-forms: an important concept, in the pages that follow—as an intrinsically expanding one.

From a single common origin, to an immense variety of solutions: it is this incessant growing-apart of life forms that the branches of a morphological tree capture with such intuitive force. ‘A tree can be viewed as a simplified description of a matrix of distances’, write Cavalli-Sforza, Menozzi and Piazza in the methodological prelude to their History and Geography of Human Genes; and figure 2, with its mirror-like alignment of genetic groups and linguistic families drifting away from each other (in a ‘correspondence [that] is remarkably high but not perfect’, as they note with aristocratic aplomb),footnote8 makes clear what they mean: a tree is a way of sketching how far a certain language has moved from another one, or from their common point of origin.

Dendrogram of a genetic tree comparing linguistic families and superfamilies published in Cavalli-Sforza et. al. in 1988. Populations pooled on the basis of linguistic classifications belong to the following groups: Bantu, Niger-Kordofanian family, Nilotic, Nilo-Saharan family, Southeast Indian, Dravidian family, Samoyeds, Uralic family from Russia, North Turkic, branch of Altaic family, Northwest Amerind, Na-Dene family. The genetic tree was constructed by average linkage analysis of Nei's genetic distances. Note from L. Luca Cavalli-Sforza, Paolo Menozzi, and Alberto Piazza from 'The history and geography of human genes' reads: why is there a close similarity between linguistic and genetic trees? The correlation is certainly not due to the effect of genes on languages, if anything, it is likely that there is a reverse influence, in that linguistic barriers may strengthen the genetic isolation between groups speaking different languages. The explanation of the parallelism between genetic and linguistic trees is to be sought in the common effect of events determining the separation of the two groups. After fission and migration of one or both moieties to a different area, they are partially or completely isolated from each other. Reciprocal isolation causes both genetic and linguistic differentiation.

And if language evolves by diverging, why not literature too?