A student of Latin, reading text with the long vowels marked, may be momentarily confused when some vowel is marked differently than he expects. But he won’t be so confused once he realizes that the experts don’t always agree on the subject. In the venerable Allen and Greennough’s Latin Grammar he will find “magnus” with the “a” marked long. But in textbooks, like Wheelock’s, the “a” is short. But sometimes the differences are only apparent, and not real. In Cassell’s New Latin Dictionary he will find many vowels not marked long which are long in other sources. However, eventually he will realize that, although the dictionary does not state so anywhere, the dictionary does not mark vowels as long, if they are long by rule. For instance, the final “o” in words is not marked as long. And neither is the “i” in insula, long according to the rule that a vowel before “ns” or “nf” is long. But on the other hand, the dictionary does not mark the first “u” in “nullus” as long. As far as I know there is no rule that would make that “u” long. So that would represent a genuine difference, as compared to other sources. The most important difference I have found is in the pronunciation of the perfect tense of the subjunctive mood. In Allen And Greennough the “i” in the endings “erimus”, “eris”, and “eritis” is short, but in other sources the “i” is long.
Have a look at Allen&Greenough’s section 169, specifically 169.d note. These endings are normally short; however, are long I is sometimes found. Although A&G don’t mention it, Gildersleeve does, i.e., Long I in the perfect subjunctive endings is found in poetry. If you read enough Latin poetry, you’re bound to encounter it. Here’s a note from Fordyce’s Catullus commentary on the famous Carmen V:
“The quantity of the -I- of -imus (and -itis)was originally long in the perfect subjunctive (an optative formation), short in the future perfect indicative (a conjunctive formation), and this distinction is maintained in Plautus. In later verse, for metrical convenience, it is disregarded: -ī- in the fut. perf., first found here in verse, is shown to have been Cicero’s normal pronunciation by his clausulae (Zielinski, Philol., Supp. ix. 772).”
Lewis & Short mark it as long:
https://logeion.uchicago.edu/nullus
In any event, isn’t the first syllable closed? That would make the -u- long by position. Have a look a A&G, Section 11.f note. However, just because it’s long by position, it may still be pronounced short unless it’s known from other evidence that the vowel should be pronounced long. A&G only mark vowels known to be long. The same can be said for magnus; however, looking in the Thesaurus Linguae Latinae, the length of -a- in magnus is shown as -ā- but is suspect.
Thanks Aetos, the information you gave about the pronunciation of the perfect of the subjunctive pretty well explains everything in that regard.
In regards to the first syllable of “nullus”, I was talking about the length of the “u” itself, not about the length of the syllable according to poetry. Actually, though, I think that my dictionary, Cassell,s, is a poor dictionary. There are better dictionaries. Probably Lewis and Short is a better dictionary. Apparently it has been put online, but I haven’t checked it out yet.
If you follow my link for nullus, it will take you to the Logeion website, which has several good dictionaries, including the L&S. My “deluxe” copy of Cassell’s has been gathering dust for years now.
Lewis & Short is still a good dictionary, but it is not always accurate when it comes to vowel length, especially when “hidden quantities” are concerned. Here’s a short and interesting article about this topic:
Have a look at how they mark magnus! I also suspect that due to the age of these dictionaries, some inscriptional evidence or some other form thereof wasn’t available at the time to make the determination one way or the other.
Thank you Laurentius. The ariticale is really helpful. 