reception does not define the original audience.

The book shows a Germanic Jewish vernacular mediating inherited scholarship for readers with differing language competence. Hebrew and Aramaic learning continued alongside Yiddish explanation.

## 13 What ancient DNA adds

Shamam Waldman and colleagues’ [2022 Erfurt study](https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/2022_Waldman_ErfurtMedievalJews_Cell%5B15%5D.pdf) obtained genome-wide data from 33 individuals in the Jewish cemetery; ten were radiocarbon-dated to about 1270–1400. The sample was more genetically varied than modern Ashkenazim, with two subgroups and evidence that the Ashkenazi founder event, a population bottleneck, pre-dated the fourteenth century.

About a third shared one maternal lineage: the paper identifies eleven of thirty-one unrelated individuals with that lineage. Disease-associated variants enriched among modern Ashkenazim were already present. These details are also highlighted in Technology Networks’ summary, “[Ancient DNA Tells the Origin Story of Ashkenazi Jews](https://www.technologynetworks.com/genomics/news/ancient-dna-tells-the-origin-story-of-ashkenazi-jews-368070)” (1 December 2022). A bottleneck concerns the restricted ancestral population from which later descendants emerged. It is not a linguistic event: a population can pass through such a restriction while retaining, adopting, or mixing different vernaculars.

Ancestry modelling gives the Italian connection a qualified form. Plausible South Italian-source models averaged about 65% southern Italian-like, 19% Middle Eastern, and 16% Eastern European ancestry. Alternative models gave ranges of 19–43% Middle Eastern and 37–65% Mediterranean European. The authors connect the South Italian fit with historical models pointing to Italy, while stressing model dependence and modern proxies. They could not reliably date the admixture.

Genomes and vocabulary supply independent evidence. An Italian-like ancestry component need not be proved by one uncertain etymology; it likewise cannot determine whether a word arrived through French, German, or later Italian contact.

“Italian-like” describes similarity to a reference population used in a model, not a language spoken by an identified ancestor. As section 1 argues, a community could retain Mediterranean ancestry while adopting a Germanic vernacular. An ancestry component does not specify the towns its bearers passed through, the sequence of their movement, or the language of each generation, and it does not show that every community followed the same route. Erfurt is one late medieval city; its internal variation argues for several community histories rather than one average applied to everyone. The genetic and linguistic layers thus answer different questions: ancestry and language do not have to line up one-to-one, as the Romance-word controls already showed.

## 14 Weinreich, Beider, and competing accounts of origin

Max Weinreich’s *History of the Yiddish Language* gives the classic large-scale account of a language formed through Jewish contact with successive surrounding vernaculars. Paul Glasser’s [YIVO account of Weinreich](https://encyclopedia.yivo.org/article.aspx/Weinreich_Max) describes the importance of Romance-speaking Jews moving into the Rhineland in this reconstruction. George Jochnowitz’s [review of the complete English edition](https://www.jochnowitz.net/Essays/HistoryYiddish.html) also explains Weinreich’s use of *Loez* for the earlier Romance-speaking cultural environments.

Alexander Beider’s [*Origins of Yiddish Dialects*](https://academic.oup.com/book/26848), published in 2015, proposes a significant revision. Its reconstruction distinguishes an East Franconian basis for westernmost Yiddish from a Bohemian German basis for Eastern Yiddish, with different small substrate components and subsequent contacts. A shared Hebrew cultural component does not, in his account, require that every variety descend from one already unified medieval Yiddish vernacular.

Beider’s relatively late date for