However great a tendency the suffixes under investigation have toward giving to the nouns a certain meaning, the variations of which they are capable,—due, as has been shown, to stem and context,—strongly suggest that there can be nothing very stable in the suffix itself. If there really were a fundamental meaning in the suffixes, there would be no such variation as we find.
But a consideration which points even more to the comparatively fluid condition of these suffixes is the fact that we find other words, formed on the same stem, but with a different suffix, meaning precisely the same as the nouns made with these suffixes. Here again, the meanings are derived from an examination of the context. Sometimes the contexts are exactly parallel, at other times there is a sufficiently large element common to both to warrant us in saying that the nouns do not, at least in these particular instances, differ in meaning.
The fact that some of these parallel words occur at different periods in the language does not weaken the argument, as the mere occurrence of them shows the unstable influence of the suffix; and, moreover, we need not suppose because one word is not found at a certain period while another on the same stem with a different suffix is found, that the first word was not in existence. It is just as reasonable to assume that the preservation of one word and not the other is due merely to common usage or the personal preference of the author. Metrical considerations might exclude the use of a certain word in poetry, but the instances are very rare, and will be noted in the proper place.
The most common suffix which makes accessory forms with -mentum is -men. Most authorities regard -mentum as an extension of -men by the addition of -to. Whether this is true or not, there are many -mentum words that have no accessory forms in -men, and a large number of -men words that have no accessory forms in -mentum. Corssen (Krit. Nach. p. 125 ff.) gives fifty-one -men words from old, classical, and later Latin to which there are no forms in -mentum, fifty-two -mentum words from the same periods to which there are no forms in -men; twenty-five words with both forms in any one period. He also gives a table showing how the words in the older and classical language preferred the form -men while in later Latin the same words preferred the form -mentum. He says the suffix -mentum is only the the extension, on Latin soil, of the suffix -men (Sanskrit, -man) with -to; and this explains why in later Latin the forms in -mentum become more frequent, also why they are not found in other Italic dialects, nor in the Greek and other related languages.
Lindsay says (p. 335) that the suffix -men is found more often in poetry, while -mentum predominates in prose.
Etymologically, the suffixes -bulum and -culum go back to original -dhlo and -tlo respectively (Lindsay pp. 334 and 332).
A study of the other suffixes which make accessory forms to these words would probably yield results similar to those seen in the case of our suffixes; but all that will be attempted here will be to show parallels wherever possible. Italics will be used here, also, to show what elements in the context go to prove the equivalence in semantic content of the nouns under discussion.
A. Parallels of -mentum and Accessory Suffixes
One of the neatest examples of identity in meaning is the following exactly parallel usage of stramen and stramentum: tectam stramine vidit casam, Ov. M. 5, 443; casae, quae stramentis tectae erant, B. G. 5, 43.
From the use of a genitive denoting a concrete object, fragmentum and fragmen are seen to be identical in meaning in the following examples: adiacebant fragmina telorum equorumque artus, Tac. A. 1, 61; tribunum adoriuntur fragmentis saeptorum, Sest. 79.