Mississippi Fred MacDowell comments on one of his own posts (using his nom de comment, "S.") that some people pronounce the biblical name Nahum as Nokhem even though the first vowel of the Hebrew name—נַחוּם—is pronounced ah in all dialects of Hebrew. This post started as a comment on the comment, but it got too long, so I made it even longer with clarifications, and here it is. Those who don't need the clarifications can skip the numbered paragraphs, but I don't know how you know whether you need them until you've read them. It might be easier to skip the numbered paragraphs for now and come back to them when you need them (if you do). There are cross-refs to these paragraphs.
1. The Hebrew vowels pataḥ ( ַ ) and ḥataf pataḥ ( ֲ ) are pronounced /a/ in both the Sefardic and the Ashkenazic dialects. The pataḥ is a short vowel, and the ḥataf pataḥ is even shorter; it should be pronounced hurriedly (see paragraph 2). The first vowel in נַחוּם is a pataḥ.
2. The ḥataf pataḥ (paragraph 1) is so short that technically it isn't even a full vowel; it doesn't produce syllables, and therefore isn't accented. It stands in for a sheva ( ְ ), which is either silent (sheva naḥ) or is pronounced like a schwa (sheva naʿ) (and this is where the English word schwa comes from). Other ḥataf vowels, which follow the same rules, are the hataf qamats ( ֳ ) and the ḥataf segol ( ֱ ).
3. There are two kinds of qamatz ( ָ ) in Hebrew; they are identical in appearance. The qamatz gadol is the long vowel corresponding to the pataḥ (paragraph 1); the qamatz qatan is the short vowel corresponding to the ḥolam ( ֹ ). (The long and short qamatzim are not the long and short vowels corresponding to each other.) Both qamatzim are pronounced /o||u/ in Ashkenazic Hebrew. In Sefardic pronunciation, the qamatz gadol (which appears much more often) is pronounced /a/, and the qamatz qatan is (or "should" be) pronounced /o/, but it usually isn't. So from a linguistic point of view, why SHOULD something be pronounced in a way most people usually don't pronounce it? Good question.
4. A dagesh is a dot appearing within a consonant, and there are two kinds. A dagesh ḥazaq may appear in any letter other than an alef, hei, ḥet, or ayin, and only rarely appears in a resh. (When a dot appears within an alef [rare occurrence] or a hei, it's a mappiq, which is a whole other story.) It serves to geminate (double the pronunciation) of the consonant; for example, הֵלּׅמּוּד is pronounced /hallimmud/, not /*halimud/. Or not; we don't really geminate in either Ashkenazic or Sefardic pronunciation. Gemination is sometimes important for theoretical purposes, some are careful about it when chanting the Torah or saying the Shema, and I believe the Yemenite Hebrew pronunciation preserves the gemination. In the case of a bet, gimmel, dalet, kaf, peh, or tav (the begad kefat letters), the dagesh ḥazaq also makes the consonant a stop instead of a fricative. The second kind of dagesh is the dagesh qal, which appears only in the begad kefats and only makes the letters into stops; it doesn't geminate. In practice, gimmel and dalet are always stops in Ashkenazic and Sefardic pronunciation, and in Sefardic tav is always a stop as well. Again, this is sometimes theoretically important even when it doesn't affect pronunciation.
5. An open syllable in Hebrew is one that ends in a vowel, and a closed syllable ends in a consonant. A consonant that is geminated with a dagesh ḥazaq (paragraph 4) or has a sheva naḥ or a ḥataf vowel standing in for a sheva naḥ (paragraph 2) closes a syllable. Broken into syllables, הֵלּׅמּוּד is /hal.lim.mud/.
Max Weinreich notes that in the Hebrew component of Yiddish, closed syllables tend to have short vowels and open syllables tend to have long vowels; although this tendency applies to the Hebrew component, it comes from the German component. He mentions several words that have either a pataḥ or a ḥataf pataḥ (/a/) (paragraph 1) in Hebrew that is pronounced like a qamatz (/o||u/) (paragraph 3). The Hebrew syllables of interest are open (paragraph 5). Recall that the qamatz gadol is the long vowel associated with pataḥ. Among the words Weinreich mentions are hodes (myrtle branch), kholem (dream), khotse (half), kadokhes (ague), tokhes (buttock), and tones (fast), as well as Nokhem.
There may be at least one mistake in the list. Lehokhes (spite), a Yiddish noun, is a repurposing of the Hebrew infinitive לְהַכְעׅיס (to anger). The sheva here is naḥ, which closes the syllable; Weinreich says this is an open syllable. He notes that in eastern Yiddish, it's pronounced lehakhes. Lehakhes is also the pronunciation given in Uriel Weinreich's dictionary.
Some of the other things Weinreich says about open syllable/long vowel and closed syllable/short vowel correspondences don't seem intuitively right to me, but I'm not chutzpadik enough to say he's wrong unless I'm sure, which I'm not.
Source: Max Weinreich, History of the Yiddish Language, 2 vols., ed. Paul Glasser, trans. Shlomo Noble with the assistance of Joshua A. Fishman (New Haven: Yale University Press, 2008); original Yiddish publication, 4 vols., 1973. The list of words is from 2008, 2: 389/1973, 2: 44; the statement about the phenomenon's German origin is from 2008, 2: 387/1973, 2: 42-43.
