I ask this since there is so much variation in this situation. For example, s after ex- prefixed words mostly gets omitted in later period texts, but can also be found in Vergil's work, despite also finding forms retaining the s. Additionally, is there development between the two forms, were the forms conflated by later editors for later editions?

  • 3
    Can you give more examples? You say that Vergil has forms both ways, but I'm not sure what forms you mean. Examples would clarify the question, at least to me.
    – Joonas Ilmavirta
    Sep 19, 2022 at 3:25

1 Answer 1


I don't know if there are any clear patterns to the historical use of the spellings ex- vs. exs-. I would expect any word to be able to show either spelling; whether actual attestations exist though would depend on chance.

The only thing I've found so far that attempts to describe ancient practice is a brief sentence in the book Aids to Latin Orthography (Wilhelm Brambach, transl. by W. Gordon McCabe, 1977) saying "the text-books in ancient times required that this S should be written after X" (in words starting with ex-s...) (page 24). I don't know what textbooks this is referring to, but my sense is that their pronouncements were probably not particularly strictly observed.

One thing to note is that it's unlikely that there was much if any difference between the sound of x and xs. So any patterns are probably best understood as spelling conventions (like the convention of writing the nominative singular forms plebs and urbs, which were pronounced with final /ps/, with the letters "b" to match the spelling of their oblique forms), not as representations of contrasting alternative pronunciations of a word.

In ancient times, XS could be an unetymological spelling of X

Latin manuscripts and inscriptions use xs not only in words where the spelling is justified by etymology (e.g. where the prefix ex- comes before a base starting in s) but also in words where it is not: spellings such as proxsume, maxsime, etc. are not rare. I'm not sure how much the absence of that kind of non-etymological xs from modern editions of major texts is due to standardization.

I'm not sure I find this hypothesis convincing, but J.N. Adams (2013) has suggested that the use of the spelling xs in words such as the above may have been intended as a way to emphasize the pronunciation /ks/ as opposed to an assimilated pronunciation /s/ (before a consonant) or /ss/ (between vowels) which may have existed as an occasional (although probably not as a systematically regular) simplification of the cluster in ancient times:

The learned spelling xs for x (for which see e.g. Väänänen 1966: 64, Adams 1995a: 90-1) might conceivably have been taught in the schools as a strategy for countering an assimilation taking place in speech. Two letters on the page prompt the articulating of two consonants.

(Social Variation and the Latin Language, page 171)

Other sources describe the use of xs in place of x as an archaic feature:

The writing of xs for x, as saxsum, belongs mostly to the Gracchan period.

(F. D. Allen (1880), Remnants of Early Latin, page 8)

spellings in <xs> for the [ks] cluster are actually archaizing. It is therefore not certain that they reflect a phonological change

(Éloïse Lemay (2017), "Studies in Merovingian Latin", page 60)

Modern normalized systems of Latin spelling may distinguish ex- from exs- based on etymology

Modern teachers, standardizers etc. seem to have come down in favor of exs- (where etymologically appropriate), so I expect some modern editions may normalize the ex- of manuscripts to exs-.

Bennett 1907 describes exsanguis etc. as "better than" exanguis etc. but doesn't explain why or what he means by this (The Latin Language, page 85).

From Recommendations of the Classical Association on the Teaching of Latin and Greek, the report on "Spelling and Printing of Latin Texts" (1905) prescribes

exs rather than ex (where possible) in compounds whose second member begins with s.

I don't understand the significance of "where possible" here (if it just meant "where etymological", it would be made redundant by the rest of the sentence).

  • 1
    "Ex-anguis" would be the natural word to use to describe a Gorgon with alopecia. Sep 21, 2022 at 7:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.