Mendele: Yiddish literature and language ______________________________________________________ Contents of Vol. 4.191 November 20, 1994 1) Spelling (Arn Abramson) 1)---------------------------------------------------- Date: Sat, 19 Nov 94 23:52:48 EST From: abramson@uconnvm.uconn.edu Subject: Spelling Since the 13th of November other pressures have kept me from commenting on the several interesting responses to my remarks about ways of writing the Yiddish "redt (ret), bistu (bist du), hostu (host du), onnemen (onemen)." Meylekh Viswanath (14 Nov.) actually joins me in expressing bewilderment over inconsistencies in applying the principles of transcription versus those of transliteration. Mikhl Herzog (13 Nov.) elaborated on the question by bringing in other examples of homophones that are spelled differently in both Yiddish script and in YIVO writing. Unfortunately, before going to the trouble of giving me undeserved credit for historical thoughts on early /bis du/ rather than /bist du/ he could not have seen my confession of an oversight (also 13 Nov.) in failing to write "bist." Actually, however, I found Mikhl's thoughts along these historical lines very interesting indeed. (He also wrung my heart for a moment with his relevant reminder of a speech mannerism of our late teacher and friend Uriel Weinreich.) I think, Mikhl, we do have to go back to our synchronic stage of the language for our discussion. (By the way, it would be interesting to know whether there was variation in scribal practices in writing all these words in Yiddish literature before the standardization of the orthography.) Phonetically, the form /bistu/, as has been pointed out by Rick Turkel (14 Nov.), represents assimilation of the final /t/ of /bist/ and the initial /d/ of /du/. This is not surprising because in casual speech maintaining of the full forms would require releasing of the articulation for the final /t/ and then making the lingual closure again to produce the initial /d/. The melding of the two homorganic (=produced at the same place of articulation) consonants yields a single consonant, which, in the context of preceding /s/, loses its aspiration; the result is neither a normal /d/ as in isolated /du/ nor a normal /t/, as in isolated /tu/, where it is aspirated (=noisy span of turbulent air before the onset of voice from the vocal folds). This resulting stop consonant is conventionally treated as "voiceless," even though there is no conmtrast between /d/ and /t/ (or /b/ and /p/ and /g/ and /k/) in this context. What bothers Meylekh and me in such spellings is that Mikhl's principle of morpheme-preservation in writing is observed in words like "redt" but not in expressions like "bistu" and "hostu." The discussion is complicated by an observational disagreement between Mikhl Herzog and Arre Komar (13 Nov.) Mikhl agrees with me that the words "redt" '(he) speaks' and "ret" '(he) rescues' are homophones (pronounced the same), which is what Meylekh and I believe. Arre, however, says, "I notice that when I speak I make a clear distinction between redt and ret. The former ending is slightly softer and more d-like than the latter." In a private message Shleyme Axelrod asked me whether such a claim could be tested by means of spectrograms. Let me give you my answer to him with, I hope, some improvement. Statements such Arre's are often made by literate folk under the influence of the orthographic vagaries of their cultures' writing systems. It would be as if some English speaker said that there is just a very small difference between "night" and "knight." (Of course, there was a very real difference beteen them, I dare say, in Old English and presumably Proto-Germanic.) If Brother Arre has fallen victim to this phenomenon and simply imagines a difference between the two words, he will have to be seized and purged of his sin. If, however, Arre is right, it follows that we have made an important discovery about the consonant system of Yiddish! There are three forms: /ret/ 'rescues,' /red/ 'speak,' and, let us say, /reD/ 'speaks.' That is, Yiddish has not only /d t/ but also some third stop consonant in contrast with both of them, that is "a little softer and more d-like" than /t/; I label his sound /D/. I said to Shleyme that before doing an instrumental investigation of the matter, we would have to determine that the three categories actually exist in the production and perception of native speakers. We would have a few native speakers record several instances each of "ikh red," "er redt," and "er ret." It is absolutely vital that they _not_ be told the purpose of the exercise. Thus, it would be best to mix a number of other short items into the list. (We would want to keep track of what dialects of Yiddish we were recording.) Removing the pronouns from the recordings, we would randomize the tokens of the key words and play them to a jury of native speakers, including our informants, for labeling in Yiddish script. If Arre is right, three statistically significant categories will arise from the data. Then we can examine sound spectrograms or other instrumental displays to determine the differentiating phonetic properties. Finally, there is the matter of the spelling of the end of one morpheme and the beginning of the next one with the same consonant letter whether or not this reflects true consonant length (gemination). Dovid Braun, who first called the matter to our attention, has added (14 Nov.),"After composing my entry on double consonants, I reread it and took a double-take, realizing that, as it stood, I was claiming what Arn Abramson says I claim. And after intuiting a bit and thinking back to the Yiddish linguistic literature I've read that touches on the matter, I was satisfied with my claim: normal spoken Yiddish has a phonological (postlexical, if you will) rule of degemination. Wherever you have two like consonants meeting, pronounce the sequence as if there were only one. "madam meyerson" is syllabified as [ma.da.me.yer.son]; "onnemen" is syllabified as [o.ne.men]. In "careful speech", a speaker might lengthen the consonant for purposes of disambiguation or phonetically preserving the integrity of each morpheme, but in casual speech, this is lost categorically." He goes on to give a note on Soviet spelling. I cannot argue with him, as my only knowledge of usage may be based on careful, deliberate speech. If he is right, of course, Meylekh and I will have to resign ourselves to further exceptions to the principles of transcription. What we seem to have in both Yiddish writing and YIVO writing is neither a pure transcription nor a pure transliteration. It is roughly a roman transcription that, with some troubling exceptions (e.g., /bistu/), preserves information about morpheme boundaries. Curiously, in Hebrew-Aramaic words Yiddish writing is normally pure transliteration, while in YIVO writng it is transcription of the pronunciation Yiddish (Semitic spelling) YIVO Gloss shin-beyz-sov shabes 'Sabbath' mem-nun-khes minkhe 'afternoon service' khes-sov-langer nun khosn 'bridegroom' Arn Abramson ______________________________________________________ End of Mendele Vol. 4.191 Mendele has 2 rules: 1. Provide a meaningful Subject: line 2. Sign your article (full name please) A Table of Contents is now available via anonymous ftp, along with weekly updates. Anonymous ftp archives available on: ftp.mendele.trincoll.edu in the directory pub/mendele/files Archives available via gopher on: gopher.cic.net Send articles to: mendele@yalevm.ycc.yale.edu Send change-of-status messages to: listserv@yalevm.ycc.yale.edu a. For a temporary stop: set mendele nomail b. To resume delivery: set mendele mail c. To unsubscribe kholile: unsub mendele Other business: nmiller@mail.trincoll.edu