*phrātēr* also shows that cognates need not retain identical social meanings.

These nouns place Latin within Indo-European, whose common inheritance is deeper than the *Aeneid*’s proposed migration. They do not locate Roman ancestors in Troy or make the Romans Greek settlers merely because Greek has cognates.

## 3 The specifically Italic layer

Michael Weiss’s chapter [“Italic”](https://www.cambridge.org/core/books/indoeuropean-languagefamily/italic/473A395FABB6E9FB91AE75076F390D29), in *The Indo-European Language Family: A Phylogenetic Perspective* (2022), places Latin within the Italic branch alongside the Sabellic languages, including Oscan and Umbrian. The closely related speech of Falerii is traditionally called Faliscan; Weiss treats its relationship to Latin as sufficiently close to describe it as a Latin dialect. Ancient Italy nevertheless contained languages outside this branch, including non-Indo-European Etruscan.

Some useful comparisons in Weiss’s discussion, supplemented by the [Osco-Umbrian languages overview](https://en.wikipedia.org/wiki/Osco-Umbrian_languages), are:

| Latin | Comparison from ancient Italy | What the comparison reveals |
|---|---|---|
| *quis*, who? | Oscan *pis*; Umbrian *pis-i* | A regular difference in the treatment of an inherited consonant |
| *vīvus*, alive | Oscan *bivus* | Related inherited material with different sound developments |
| *hortus*, garden | Oscan *húrz* | A noun related through changes in sounds and word shape |
| *medius*, middle | Oscan *mefiai* | Inherited aspirates become *b/d* inside Latin words but *f* in Sabellic |

Weiss dates the earliest Roman Latin material to the seventh–sixth centuries BCE and literary Latin’s beginnings to the third. The Lapis Satricanus dedication from Satricum, dated to the last quarter of the sixth century BCE, names a Publius Valerius. The [University of Amsterdam excavators](https://www.uva.nl/en/discipline/archaeology/research/satricum/epigraphic-finds/epigraphic-finds.html) call it the oldest coherent public monumental Latin inscription: evidence independent of Virgil.

One word can travel; recurring correspondences, grammar, and shared innovations identify descent. The pattern groups Latin with Italian relatives rather than demonstrating transplantation by Trojan founders.

This identifies the language’s ancestry, not a prehistoric passenger list. Italic-speaking populations had their own histories of movement, and newcomers could adopt their language.

Greek literature does not establish what historical Trojans spoke. Bronze Age Troy’s language is unknown; a [single portable Hieroglyphic Luwian seal from Troy VIIb](https://en.wikipedia.org/wiki/Trojan_language) is inconclusive. The test cannot simply be whether Latin is Luwian, but whether any specifically Trojan residue remains unexplained by the Italic account and known contacts. A Trojan-language hypothesis needs diagnostic evidence; broad Indo-European similarities would not suffice.

## 4 Farming, herding, and food production

Much subsistence vocabulary continues the inherited pattern.

Latin [*ager*](https://alatius.com/ls/index.php?l=ager), field or territory, belongs beside Greek *agros* and English *acre*. Latin *ovis*, sheep, and *porcus*, pig, likewise belong to inherited Indo-European vocabulary. Such words establish that Latin speakers did not need to obtain their entire language of farming and herding from a late-arriving eastern founder.

These words do not imply that their speakers invented agriculture or remained on the same land. Inherited nouns survive migrations and changes of meaning: English *acre* became a unit of area, whereas Latin *ager* could denote a field, territory, or district.

Some Latin farm and food vocabulary nevertheless shows the sound changes of Sabellic-speaking neighbours. The standard handbook view, represented by Michiel de Vaan’s *Etymological Dictionary of Latin and the Other Italic Languages* (2008), under *bōs*, *rūfus*, and *popīna*, treats these forms as evidence of borrowing within Italy. Roman tradition itself made