Who first drew a family tree as a tree?

The Online Etymology Dictionary indicates that the English-language expression "Family tree" in the sense of "graph of ancestral relations" is first attested from 1752, in the novel A Genuine Account of the Life and Transactions of Howell ap David Price (which is available in Google Books).

Such pedigree diagrams have a much longer history, of course, but they were not called family trees, nor were they drawn with any particular tree-like imagery (except for the religious Tree of Jesse, pictures of which started appearing in the 10th century). See, for example:

• The role of biblical genealogies in phylogenetics
• The first infographic was a genealogy (c. 400 AD)
• How confusing were the first written genealogies?
• The first royal pedigree
• Another early noble pedigree
• The first known pedigree of a non-noble family

This leaves open the question of who first drew a tree-like family tree. [Note: see also the later post Drawing family trees as trees.]

Ernest H. Wilkins (1925. The genealogy of the genealogical trees of the Genealogia deorum. Modern Philology 23: 61-65) has suggested that it might be the Italian author and poet Giovanni Boccaccio (1313-1375), in his Genealogia Deorum Gentilium (On the Genealogy of the Gods of the Gentiles).

This Renaissance book was an "encyclopedic compilation of the tangled family relationships of the classical pantheons of Ancient Greece and Rome" (according to Wikipedia). It was written in Latin, apparently starting in c. 1350, and then continuously corrected and revised until the author's death. In c. 1370 an apograph [ie. perfect copy] was made of an autograph manuscript [ie. in the author's own hand], and from that first apograph other copies were made.

The 1370 autograph is not known to still exist; but a second autograph manuscript, showing later revisions, is in the Laurentian Library in Florence (MS. LII, 9). There are some three dozen extant apographs from the 1300s and 1400s, all based on the lost first autograph. The first printed edition was produced in Venice in 1472, followed by an edition of 1473 printed in Leuven. At least seven other editions appeared during the 1400s and 1500s. A French translation was published in Paris in 1498, and an Italian translation appeared in Venice in 1547. (See Ernest H. Wilkins. 1919. The genealogy of the editions of the Genealogia Deorum. Modern Philology 17: 425-438.)

The illustrations shown here are from various versions of the book.

Wilkins (1925) notes:

The extant autograph manuscript of the Genealogia Deorum of Boccaccio is illustrated by thirteen genealogical trees, designed certainly and drawn in all probability by Boccaccio himself. At the top of each tree is a large circle, in which is written the name of a divinity. From this circle descends a stem which now expands into other lesser circles, now sends forth leaves, and now branches, which in their turn expand into circles and send forth leaves and lesser branches. In the center of each circle or leaf a name is written. The circles are used for those divinities whose progeny is represented in the same tree; the leaves, for divinities whose progeny is not represented. In the circles the words qui genuit [ie. who fathered] follow each masculine name, and the words quae peperit [ie. who bore] each feminine name. Similar trees certainly appeared in the earlier lost autograph, from which all the apograph manuscripts are derived; and similar trees appear in several apographs, and in the fourth and all later editions of the Genealogia.

So far as I can ascertain, Boccaccio's trees are the earliest secular genealogical trees properly so called: that is to say, the first non-biblical genealogical charts in which stems, branches, and leaves appear.

This claim of priority has apparently gone unchallenged by later workers; eg. Christiane Klapisch-Zuber (1991. The genesis of the family tree. I Tatti Studies in the Italian Renaissance 4: 105-129) notes:

It may well be that Boccaccio was the first to combine the old graphic system of medallions in the descending order typical of medieval genealogies with the implications of a vegetal theme.

The vegetal image is quite obvious, although the leaves do vary widely in form within any one manuscript, and also from copy to copy. In the autograph they are palmately five-lobed. In some trees the different generations are indicated by variation in the colour of the branches.

Personally, to me each of these diagrams looks more like a vine than a tree, especially with the root at the top.

Moreover, some of the printed editions do not contain the genealogies, and in others their form is modified. For example, some have a portrait of the progenitor divinity, and others bear scrolls or circles instead of leaves. Some of the trees have extra (empty) leaves or scrolls. It is thus quite clear that the tree metaphor for the pedigrees was not seen as important at the time.

Nevertheless, it is important to note that in the first two editions of the Italian translation by Giuseppe Betussi (1547 & 1554; but not in later editions) the first genealogy is drawn as an actual tree rooted in the ground, with the name of the progenitor appearing at the base of the trunk. Klapisch-Zuber notes:

In comparison with Boccaccio's divinely radiant foliage, this image must strike us as mean and desiccated. And yet, it is the triumph of the genealogical tree as we know it, planted right side up; and any one in the modern world can use it to evoke his ancestors and to express his faith in the survival of his lineage.