Talk:List of glossing abbreviations

Apodosis - APOD?
Protasis (PROT) is present, so would apodosis have the glossing abbreviation (APOD)? Jackwolfroven (talk) 03:06, 7 July 2013 (UTC)


 * Haven't seen it, but that would be reasonable. — kwami (talk) 03:20, 7 July 2013 (UTC)

"Zoic Gender"?
The meaning of "zo" is given as "zoic gender (animals)", but the link that "zoic gender" leads to has no mention of this phrase. A quick internet search gave me no explanation for the term either. Can this description be elaborated on or changed to be more clear? Fench (talk) 22:37, 12 June 2014 (UTC)

ABSV - absentive constructions, i.e. those conveying the meaning of an event occuring in a place displaced from the deictic centre (Bertinetto, Pier Marco/ Ebert, Karen H./ de Groot, Casper. The progressive in Europe. In: Tense and Aspect in the Languages of Europe. Dahl, Östen (Ed.)


 * Thanks. Added. — kwami (talk) 19:07, 1 June 2021 (UTC)

More items needed
There are many glossing abbreviations in articles on Wikipedia which do not appear in this list, creating little mysteries for the average reader. Please make it a custom that, whenever a glossing abbreviation is used, this list should be checked, and edited if needed.   Solo Owl   00:23, 13 May 2018 (UTC)
 * There's quite a lot of variation out there, and one and the same abbreviation can be used by different authors to mean different things, so keeping a standardised list like this is probably not the optimal solution. Ideally, abbreviations should be annotated with their meaning whenever they're used, say like that: ACC (the meaning appears in a tooltip on mouseover). If all abbreviations in an article are properly formatted (with gcl as in this example, or – if found within interlinear text – using interlinear), then they will be no need for readers to look them up here. – Uanfala (talk) 08:46, 13 May 2018 (UTC)
 * Whether optimal or not, this is an extremely useful page, and I and my students reference it often. Tooltips are all well and good, but when reading linguistics papers without a glossing key, this list is a lifesaver. As much as other wikipedia articles could be better formatted, there is a world outside of wikipedia, and the need for a page of common glosses like this is not to be understated. I also wonder whether the added column of references is that useful. Many of these glosses don't seem to have a definitive source, but have become common through repeated usage. Does that mean any example of usage will do? – Drdrphd (talk) 16:29, 23 July 2020 (UTC)
 * I fully agree on the usefulness. A practical problem is that the same gloss can have different meanings, and that a single meaning can have multiple glosses. I suggest the following approach: (1) Record as many gloss variants as possible to make the list useful for readers, but (2) mark at most one variant as the conventional one to make the list useful for people who want to do glossing. The "conventional" gloss may eventually be the first version a literature reference will be provided for ;) I revised the table structure accordingly, and marked Leipzig glosses as conventional (wherever applicable), other glosses as conventional if a source was given and only one variant was listed. Chiarcos (talk) 02:20, 8 August 2020 (UTC)

Movement = MVT or MVMT?
Both forms appear in the list, MVT directly and MVMT as an equivalent for LAT. Who uses MVMT? yoyo (talk) 16:12, 4 September 2019 (UTC)


 * Of course, it doesn't matter who, just that it's used somewhere.  thinking about what you wrote just above: "… keeping a standardised list like this is probably not the optimal solution", and seeing that quite a few different glossing abbreviations have identical expansions, should we be keeping a list of standardised meanings with variant indices, with each meaning having a standard (unique) expansion, and (usually) a wikilinked grammar article?  That is, a two-stage lookup e.g.

Stage 1: … LAT -> … MVMT -> … MVT -> … NMZ -> … NOMI -> … NZ -> … Stage 2: … -> lative case (= MVMT, direction) … -> nominalizer/nominalization …

This approach would give us consistent translations for each gloss.

However, we have a one-many relation between abbreviations and (linguistic) meanings. E.g. from the list: … MOD -> mood, modal, modal case …

Could we model this one-many relationship effectively in this two-stage approach, thus?:

Stage 1: … MOD ->, , … Stage 2: … -> mood … -> modal … -> modal case …

The second stage would still have to be one-one, since it succinctly represents the meaning in a rather abstract way. The whole meaning may take a very large WP article to express and explain, so the expansion we produce is just a kind of hint, for the reader to follow up if they need to. But this hint needs to point to just one meaning; so if necessary, the expansion may be a list of hints to the variant meanings of the gloss.

AFAIK, the only way to effectively decompose a many-many relation — as we have here between glossing abbreviations ("glosses") and linguistic meanings ("translations") — is by interpolating an entity between them ("conjunction") so that we create two relations, each many-one. After all, the glosses are abbreviations, and therefore may lose information. Logically, we start with the meanings, which we label with terms such as "grammatical mood"; in practice, we abbreviate the phrase to simply "mood", relying on context to inform the hearer or reader that we mean grammatical mood rather than temperamental mood, say; then in glossing, we choose a few letters to code that meaning, say "MOD". The meaning precedes the gloss; the gloss is a pointer to the meaning, but an imperfect one.

Perhaps my terminology is muddled! To start with, I propose these terms (but feel free to improve on them!): meaning = one specific grammatical term or function; hint = a short form for a meaning; one or more hints may point to just one meaning; gloss = an abbreviation for grammatical terms or functions; a gloss may point to one or more hints

Thoughts? yoyo (talk) 05:42, 5 September 2019 (UTC)
 * This article is a list of abbreviations as they are used in the wider world by various linguists in various ways for various languages. I don't think we can do much about the way it's organised, but I take it that what you're referring to is the way glossing abbreviations are used within articles here on wikipedia. A simplified, maybe too simplified, form of matching glosses to hints/meanings is done by the templates (gcl and interlinear). If what they do is to be made more sophisticated, then the process you describe is the way to go. However, I don't know how far these teamplates should in attempting to supply default meanings for the abbreviation used by editors; I've come to see this functionality simply as a bootstrapping trick: a device to make it easier to add glosses that look acceptable, but one that we would ideally not rely on in the long run. Ideally, the meaning of all glosses used in a given article should be explicitly set within the article (as is done for example here). The next step could be to standardise, within limits, the abbreviations used across articles. We can't do much about hints/meanings (too dependent on the language or the analysis), but maybe we could standardise the glosses, for example making sure "nominaliser" is always abbreviated as NMLZ, rather than NMZ or NZ.
 * Anyway, I digress. Have I correctly understood the context of your proposal? Are you thinking of a way to have the hints be represented differently than the meanings? – Uanfala (talk) 20:08, 7 September 2019 (UTC)

Abbreviations for content words (was: What about noun?)
This page offers numerous abbreviations, but the one for the nouns is missing ... where can I find this information? Can anyone add it? Francesco Miracapillo (talk) 16:52, 2 January 2021 (UTC)


 * In interlinear glossing, content words are normally not abbreviated with a tag, but glossed with their translation, e.g., one would gloss German *Häuser* "houses" as *house-PL* (or the like) not as *NOUN-PL*. So, there is no abbreviation for noun *in this list*. However, you might want to look into the UniMorph specifications, these are inspired by *both* glossing conventions and part-of-speech tagging, and out of the latter tradition, they also introduced abbreviations for content words, e.g., `N` for nouns. Chiarcos (talk) 09:17, 5 January 2021 (UTC)

Internal consolidation
At the moment, the conventional glosses on the list of glossing abbreviations are selected based on an (implicit) ranking of literature references, with Leipzig Glossing Rules taking priority over others (in their order of consultation). The resulting selection of conventional glosses is not always consistent, as different sources apply different rules for creating abbreviations, e.g., there is a rule to abbreviate XY-essive as -E (original Wikipedia rule) or as -ESS (implicit rule in Lehmann 2004). Likewise, the N-rule (not-XY) is not systematically applied, and often used along with direct abbreviations; same for the -izer (-Z) rule. Similarly, rules for abbreviating Anti- (A- or ANTI-), Pre-/Post- (P. or PRE-/POST-), Pro- (mostly PRO-), -lative (-LAT or -L), etc. We need a discussion about such rules and their priorities. The problem is that they are usually implicit in the literature.

Such rules, and their priorities should be discussed here. In case of doubt, we should follow established Wikipedia practice if it can be shown to follow any existing practice in linguistics. Chiarcos (talk) 13:45, 10 August 2020 (UTC)


 * I agree that we should avoid ambiguity in WP glossing as much as possible, and that our choices of preferred glosses should reflect that. However, I disagree that there should be only one preferred gloss per term. Glossing should take the audience into account. If a gloss only occurs once or twice, it may be better to use a more transparent gloss than in the Leipzig list, or even to not abbreviate it at all but only set the grammatical term in small caps. That's the same rule of thumb as not introducing acronyms into our writing if we're not actually going to use them, or if we only use them once or twice so the reader doesn't have a chance to get accustomed to them and they only hinder comprehension.
 * On the other hand, if a gloss is highly frequent in a text, and its length is causing problems with aligning the gloss to the text (leaving huge gaps in the top line), it may make sense to use a shorter abbreviation. Two examples I can think of are inclusive/exclusive, which are commonly reduced from and  to  and, and antipassive/anticausative, which are normally  and  (or  and  or something similarly long) because they're not familiar to most people, but may be just  and  in texts where they're ubiquitous, the reader can be expected to become familiar with them, and using the longer forms would be disruptive. On the other hand, if they only occur once or twice in a text aimed at a naive audience, it might be better to use  and  so the reader isn't forced to check the list of abbreviations.
 * Leipzig captured something of this practical approach in their optional rule for reducing PNG marking, eg. to $⟨3ms⟩$, but it's useful elsewhere as well.
 * It's a different matter when there are half a dozen abbreviations for the same term, with little difference in length, or when different terms are used for the same grammatical form. There I agree that selecting one, as Leipzig did, is beneficial to the reader (and to the writer, for that matter). IMO, we should prefer abbreviations that are relatively (a) short [3 letters if possible], (b) recognizable and (c) unambiguous, and terms (what the glosses are abbreviations of) that are (a) recognizable and (b) not likely to be confused with another. It's a great help when authors use for subjunctive and  for perfective, and it's annoying to read older texts that use ambiguous glosses. — kwami (talk) 19:39, 1 June 2021 (UTC)

I just found an introductory source that gave just this advice, with the example of 'downriver' being glossed as or  depending on the needs of the author. — kwami (talk) 01:15, 2 June 2021 (UTC)


 * I agree on the need to have more than one term per gloss for linguistic glossing, but https://en.wikipedia.org/wiki/Template:Interlinear will require unambiguous glosses in order to be effective (i.e., to point to the definition). We have to find a compromise between Wikipedia features and linguistic practice (which should allow for much more flexibility). Chiarcos (talk) 15:18, 22 June 2021 (UTC)


 * The glosses can be defined manually, but why can the common doubles (e.g. ANTIC and AC or EXCL and EX) not be both defined in the template list?
 * They can, indeed. The problem is less one of implementation, but how to maintain it, especially short glosses will tend to be ambiguous. As for EX, for example, this could also mean "existential" (EX is, indeed, the POS tag for existential there in Penn Treebank), and the more glosses are in the overall list, the more likely are undetected doublets (i.e., incorrect information). But it's just fine to use a non-preferred gloss. This only means that it won't give you a mouseover event, and at the first occasion within an article, one can give both the short and long version (maybe in brackets) to clarify what is meant. Chiarcos (talk) 14:15, 30 August 2021 (UTC)
 * BTW, nice work on the template. It's really handy. — kwami (talk) 18:43, 22 June 2021 (UTC)
 * Talking about achievements: I just looked at the history. You put a *lot of effort* into this. Really impressive ;) Chiarcos (talk) 16:50, 17 September 2021 (UTC)
 * I've still got a few open tabs on my browser for additional lists of abbreviations at GBooks. I'm just a bit burned out. Maybe I'll get back to this in a couple months. — kwami (talk) 01:58, 18 September 2021 (UTC)

Polypersonal agreement
The list features the notation > for the direction of transitivity. We don't know where this comes from, but I wonder whether this is actually adequate. For polypersonal agreement with more than 2 participants, how should that be encoded with this notation? Cf. Basque

d-akar-ki-o-gu

it-bring-DAT-he-we

'We bring it to him/her'

(example from Polypersonal_agreement) Chiarcos (talk) 10:42, 10 August 2020 (UTC)


 * It wouldn't be used for Basque, because they're separate morphemes and therefore glossed separately. Rather, you'd gloss each of the three with their case or role. In a language where they're identical in form, distinguished only by order, you might add the role in parentheses as a covert parameter.
 * Each hyphenated morpheme in the text requires a separate hyphenated gloss, as a basic principle of the Leipzig rules. The only exceptions are that clitics may (optionally) be marked off with dbl hyphens (equal signs), while reduplication require a tilde and infixes angle brackets instead of hyphens. But otherwise the punctuation marks are all alternatives for the period.
 * As for whether the convention is adequate, I believe so. I've never heard of a language that encodes three persons in a single morpheme. If more were needed, even if you just didn't feel like segmenting the text, you could gloss it something like 3(T);2(R);1(A) or 1:2:3 [I (give) it to you]. It might be worth exploring ways to abbreviate that further, if the glosses are too long, maybe something like 1:3>2 (1st person acting on 3rd person for the benefit of 2nd person). — kwami (talk) 18:59, 1 June 2021 (UTC)

"cn?" tags
I've included refs for all alt forms that I've been adding over the last few days, and where I've removed the cn tags. But I didn't think to verify that the existing alt forms were in the refs. I've seen most of them, so cluttering the list with actual cn tags seems overkill, but they're not necessarily in one of the refs for their entry. (E.g. the 3 entries with alt forms ASS, maybe none ref'd, or the 3 abbr. for 'incompletive' with 2 refs.) So I've marked them by hand as "cn?". We should eventually go do a more thorough search in the basic refs like ELL2. — kwami (talk) 01:28, 5 June 2021 (UTC)

CON
CON, in the list, denoting "concrete" (as opposed to "abstract") is marked above as "citation needed". Later this year I will publish a volume of essays about translations of Jabberwocky into some 50 languages. In the essay on the (my) translation into Blissymbolics a diacritical mark called the "thing indicator" (which turns for instance a character that looks like ♡ meaning "feeling" into a different word meaning "heart") will be represented in a morphemic transcription with CON. So there will soon be a publication making use of this abbreviation. -- Evertype·✆ 20:40, 7 May 2022 (UTC)

Kinship terms
Does anybody use abbreviations for "partner" (as opposed to "spouse")? -- Evertype·✆ 19:05, 13 May 2022 (UTC)


 * I've never come across it, though that's not saying much.
 * Are there languages that have different kinship terms for "partner" and "spouse"? In societies with traditional same-sex marriage, e.g. throughout Africa, AFAIK the usual words for "spouse" are used.
 * I think the recent use of "partner" in English is a bit of an aberration, like 'acoustic guitar'. These relationships are marriages by the pre–organised religion/bureaucratic conception of 'marriage' -- when two people move in together, they're married, and when they move out, they're divorced. There is no distinction between shacking up and getting married, or between separating and getting divorced. likewise, i wouldn't expect to need distinct terms for boy/girlfriend and fiance, but it would be interesting to be wrong. — kwami (talk) 02:45, 14 May 2022 (UTC)
 * Blissymbols Has a full set of characters for MAN, WOMAN, PERSON; FATHER, MOTHER, PARENT; BROTHER, SISTER, SIBLING; HUSBAND, WIFE, PARTNER; SON, DAUGHTER, OFFSPRING, and these are all combined in pairs and in a morphemic transcription of Bliss abbreviations are likely to be useful for brevity. -- Evertype·✆ 21:51, 14 August 2022 (UTC)
 * In that case the solution is to create and define your own abbreviations. — kwami (talk) 00:32, 15 August 2022 (UTC)