Choosing -ter or -iter for adverbs from third declension adjectives

Question

The typical suffix to derive an adverb from a third declension adjective is -iter, but sometimes the -i- is dropped: dulciter but audacter. I am not asking for a rule for choosing -iter or -ter — that should be treated in a separate question if there is interest — but how far I can bend that rule.

I occasionally write metric poetry in Latin, and it is very convenient to have some rhythmic flexibility. Can I freely choose between -ter and -iter to save the scansion? I think most adjectives have a preference for one or the other, but did ancient Romans ever use both endings with the same adjective? If they did, was this more common in poetry than prose?

To give a concrete example, I could use the phrases audaciter ille canit and ille celerter abit in hexameter, but not if I change to the more typical versions audacter and celeriter.

If the stem of the adjective ends in -nt-, then the ending is -er (for example prudenter), and I see no room for movement here. Prudentiter would sound weird, although it probably would be understood correctly.

Answer 1

I have run a quick analysis using data from latinlexicon.org. I included adverbs ending in -ter (about 820). Most end in -iter (the rule). A good number end in -nter (which as you know are formed with syncopation regularly when the base adjective/participle ends in -nt). The remainders are a mix of declinables and indeclinables (e.g., propter is indeclinable so I excluded it). There were only a very few left, which can be summarized into three observations:

1. Based on an adjectives ending in -t (which does happen to include the -nt participles): there is a syncopation that seems to suggest the rule is -titer = -ter.
2. Based on an adjectives ending in -c (the only example seems to be audacter): there is a syncopation that seems to suggest the rule is -citer = cter.
3. Based on an adjectives ending in -lis (facil, dificil, simil)

The next question becomes, are there a large number of adjectives in these categories that don't conform to the preceding observations? This is a difficult question to answer because there are so few adverbs that were recorded in Latin literature.

However, here are my anecdotal notes on the three categories of adjectives:

1. This group is almost totally devoid of adverbs that haven't already been mentioned, but based on the production of -nter adverbs I would guess that it would hold true to a native Latin speaker. Honestly, most of the adjectives are place names anyway that rarely form adverbs. However, when they do form adjectives, the often use -e instead of -iter (most likely an old locative).
2. Audacter seems to be the one exception to this rule. Most other adverbs seems to comfortably end in -citer not -cter (e.g., pugnaciter).
3. These seem to comfortably form -iter adverbs (visibiliter, gentiliter). It seems like the only ones that form -ter adverbs are some of the "irregular" adjectives facil, difficil, simul, etc. (the ones that form their superlatives with -limus).

So, to answer your question with a mix of analysis and opinion, I would say only in one case would you want to syncopate, that being adjectives of the 3rd declension ending in -t (which definitely includes -nt) but not -c and not -l.

It is, of course, poetry. So I'm quite sure you could add and subtract an 'i' here and there and only the most picky editors would care. After all, Adele sings the word "hello" with 8 syllables.

Answer 2

It looks as though there are three stages:

First: inherited adverbs ending in -ter, like praeter, subter, propter (IE -ter-).

Second: the ending -ter is abstracted from these and attached to the stem of an adjective, as in brevi-ter, gravi-ter, audāc-ter.

Third: the ending -iter is abstracted from the second type and attached to the bare root of an adjective, as in firm-iter, perhaps by analogy to words like firmi-tas, itself by analogy to words like civi-tas, where the -i- is etymologically justified.