Linguistic Miscellany Thread

[bradrn]

This may sound silly, but I am still somewhat confused about how the perfective aspect works. If the perfective marks verbs as completed and considered in their entirety, does that mean it generally implies past occurrence even in languages without grammatical tense marking? What about future-tense verbs and imperatives? The verb has not yet occurred, making it incomplete, yet it refers to an event that will eventually complete.

Firstly, you’re not alone here. I was confused by the perfective for a very long time before I eventually figured it out.

Now: when reasoning about the imperfective/perfective distinction, I have found it useful to realise that these concepts behave as prototypes. The most prototypically ‘perfective’ events are defined by a cluster of features, as stated by Dahl (1985):

A PFV verb will typically denote a single event, seen as an unanalysed whole, with a well-defined result or end-state, located in the past. More often than not, the event will be punctual, or at least, it will be seen as a single transition from one state to its opposite, the duration of which can be disregarded.

However, no language has a verbal category corresponding perfectly to this definition: each language has its own rules for when certain aspectual marking is allowed or disallowed. (The way I think of it: prototypically perfective events, according to the definition, will have perfective marking in all languages; similarly for prototypically imperfective events. Languages differ in how they divide up the space in between the prototypes.) There may well be some languages in which perfective verbs are always located in the past, but rather more allow non-past events to be marked perfectively if they satisfy most of the other requirements. English is rather permissive in this regard, in that nearly any verb may be marked as ‘perfective’ (a bad term, I prefer ‘simple’ for English) as long as the relevant event is not the total opposite of Dahl’s definition.

[Richard W]

Second... there are languages where the distinction works in all three tenses. You've been speaking one all your life, in fact. So to answer your question, think about what it means to say "I will read the book" vs. "I will be reading."

Or "I will own my own house one day". What Zompist appears to be calling the perfective may be syntactically a perfective, but that doesn't seem helpful for cross-language comparisons.

[zompist]

Second... there are languages where the distinction works in all three tenses. You've been speaking one all your life, in fact. So to answer your question, think about what it means to say "I will read the book" vs. "I will be reading."

Or "I will own my own house one day". What Zompist appears to be calling the perfective may be syntactically a perfective, but that doesn't seem helpful for cross-language comparisons.

It is however helpful in answering the question that was asked. It isn't helpful to snow malloc with unexplained complications.

What we have in English is not so much perfective vs. imperfective, as unmarked vs. imperfective. However, "I will read the book tomorrow" is perfective, and most importantly, shows that we can talk about something as an event or change of state, and a completed one, in the future. Quechua works the same way: verbs can be either unmarked for aspect or explicitly imperfective, in any tense.

At the same time, malloc's worry that the perfective implies past tense is not entirely wrong either. Some languages do make the perfective vs. imperfective distinction only in the past. Semantically, we can only be sure an event is complete, or that it happened at all, if it's in the past.

[Jonlang]

Sorry to break the flow but this doesn't warrant a new thread. Is there a go-to online Irish dictionary? Like, the OED or GPC for Irish?

[Richard W]

At the same time, malloc's worry that the perfective implies past tense is not entirely wrong either. Some languages do make the perfective vs. imperfective distinction only in the past. Semantically, we can only be sure an event is complete, or that it happened at all, if it's in the past.

On the other hand, Russian uses the same inflection markings for the present of imperfective verbs and the future of perfective verbs. (I don't know what happens with declarative acts, such as 'I pronounce you man and wife'.) I'm identifying the tenses semantically, rather than by form. There may be all sorts of complications when one looks at Russian closely.

[hwhatting]

(I don't know what happens with declarative acts, such as 'I pronounce you man and wife'.)

In that sentence, Russian uses the imperfective present tense. Perfective present tense would refer to a future event. Basically, the Russian perfective is always non-present.

[Ephraim]

This may sound silly, but I am still somewhat confused about how the perfective aspect works. If the perfective marks verbs as completed and considered in their entirety, does that mean it generally implies past occurrence even in languages without grammatical tense marking? What about future-tense verbs and imperatives? The verb has not yet occurred, making it incomplete, yet it refers to an event that will eventually complete.

Not completed at the moment of speech, necessarily, but rather completed within the time referred to. This is fully compatible with both past and future time reference (although, as has been pointed out, many languages do restrict the opposition to the past tense). Also, language may differ in whether the perfective requires the event to be completed in the sense of having reached its natural endpoint (which requires the event to have such an endpoint in the first place), or whether it’s enough for the event to have stopped.

Present perfectives are another story. Since the present is basically a single moving point in time, it’s hard to fit a complete durative situation into the present time. For this reason, the present tense and perfective aspect (in their strictest senses) are somewhat semantically incompatible. This is sometimes referred to as the ’present perfective paradox’ (The Present Perfective Paradox across Languages is also the title of a 2017 book by Astrid De Witt).

Nevertheless, there are some candidates for ’true’ present perfective uses, most notably performatives (what Richard W calls “declarative acts” above) and reportive presents, i.e. for reporting events that are observed at the moment of speech, as in sports commentary.

As can be seen on table 3.2 of Dahl 1985 p. 71 f, languages differ on whether they use the perfective or the imperfective for reportive presents.

(I don't know what happens with declarative acts, such as 'I pronounce you man and wife'.)

In that sentence, Russian uses the imperfective present tense. Perfective present tense would refer to a future event. Basically, the Russian perfective is always non-present.

Apparently, Slovene normally uses the perfective presents for performatives. It seems that most other Slavic languages agree with Russian in mostly using imperfectives here, but they may have perfective presents in some other cases.
https://www.degruyter.com/document/doi/ ... -0018/html

[Moose-tache]

Uh... am I the only one who read malloc's question as being about the perfect rather than the perfective? Based on his confusion about a completed action being automatically anterior, it seems he might have been conflating the two concepts (which is sadly inevitable given the similar names).

[Kuchigakatai]

A bit over two years ago I posted:

As somebody who studies Latin, I'm annoyed by a number of unetymological spellings in Spanish that break the general rules:

- ayer 'yesterday' should be ahier, cf. its etymology ad herī
- ahí 'there' should be aí, cf. its etymology ad ibī
- aun/aún 'even [if, when...]' / 'still' should be ahun/ahún, cf. their etymology adhūc (reinterpreted later as *ad hunc)
- invierno 'winter' should be himbierno, cf. its etymology tempus hībernum
- ahora 'now' should be hahora, cf. its etymology hāc hōrā
- hombro 'shoulder' should be ombro, cf. its etymology umerus
- húmedo 'humid' should be úmedo, cf. its etymology ūmidus

The latter two are more understandable since in Medieval Latin they could perfectly be spelled humerus and hūmidus though. See also, from English:

- "to hallucinate" should be "to alucinate", cf. its etymology ālūcinārī (modern Spanish has alucinar)
- "ability" should be "hability", cf. its etymology habilitās (modern Spanish has habilidad)

Since then, I've learned of a few more:

- olvidar 'to forget' should be olbidar, cf. its etymology *oblīt-āre, derived from oblīvīscor oblītum
- basura 'trash' should be vasura, cf. its etymology *vers-ūra, derived from verrō verrere verrī versum
- boda 'wedding' should be voda, cf. its etymology vōta, plural of vōtum
- barbecho 'fallow, ploughed but unseeded ground' should be varvecho, cf. its etymology vervāctum
- hinchar 'to inflate' should be inchar, cf. its Latin etymology īnflāre, although the spelling does continue Old Spanish finchar>hinchar
- hallar 'to find' should be allar, cf. its Latin etymology afflāre, although the spelling does continue Old Spanish fallar>hallar
- coger 'to grab; to fuck' should probably be cojer: the spelling suggests it's from Latin cōgō cōgere, but it's actually from colligō colligere — I am of two minds regarding whether it should really be cojer to reflect the palatal [ʎ] it must've had in the proto-Ibero-Romance stage though, or whether alluding to the g in colligere is good enough

[Sol717]

See also, from English:

- "to hallucinate" should be "to alucinate", cf. its etymology ālūcinārī (modern Spanish has alucinar)
- "ability" should be "hability", cf. its etymology habilitās (modern Spanish has habilidad)

There's probably many more English examples of such pseudo-etymological spellings/pronunciations of Latinate words. Because distortions of words belonging to a learned stratum are obviously bound to crop up in any language with one due to the normal processes of linguistic transmission and the failures of reformers to correct them in every case, I'm not particularly annoyed by them (it is telling that I think the reform of Spanish orthography upon Latinate etymological lines should've been avoided), but I'll share a few I've noticed:

• author, authority should be either auctor, auctority after Latin or autor, autority after Old French and Middle English (a pronunciation of authority with /t/ survived into the 18th century, but was proscribed).
• Unetymological /θ/ also occurs in anthem (ultimately from Latin antiphōna; in Middle English it was far more prevalent (e.g. rethorik "rhetoric") as <th> represented /t/ in Greco-Latinate words (their ModE pronunciations with /θ/ are later developments due to a combination of etymologising and spelling pronunciation).
• inchoate should be incohate (L. incohātus; in Medieval Latin apparently often inchoātus)

[Rounin Ryuuji]

There are some of these orthographical etymological mismatches I've pondered before (specifically with regard to English):

• The endings -ity and -ty, where derived from words of French origin, should probably be -itee — abilitee, bountee, cittee (archaically, the word also had double "t", probably to indicate the first vowel was short), and so on;
• The word delight should be delite (cf. Chaucer On bokes for to rede I me delyte.);
• The ending -tion was borrowed from Norman as -cioun, and should probably be -cion or -ciun; English tends to prefer c to t for this /t/-derived s-sound elsewhere, note democracy (not *democraty), and ancient (not antient, though this spelling did persist through some of the Nineteenth Century);
• The word choir has an archaic spelling quire, but could also be appropriately quiar, note that it follows a similar phonetic development, from Norman quere, as friar from frere, and also native briar from brere;

A few other points about English spelling and how I would expect it to have developed, even though it didn't:

• Middle English /aː ai/ appear to have merged, so I would expect many instances of formerly long /aː/ to be orthographically ai, ay, possibly with a terminal -e as a nod to their previous a + consonant + e spelling: pane > *paine, cane > *caine, wane > *waine;
• The newly-emerged /ɑː/, I would expect to become the new default value of "long a"; this would affect mostly recent borrowings;
• As long e is now /iː/, I would expect long o to be /uː/, and ea to be /eː/, as oa is /oː/ — speak > *speke, but break, great remain unchanged; choke > choak (encountered through the Nineteenth Century), but to, too are unchanged;
• Middle English y being used often for [iː], I would expect it to become a default spelling of [ai], especially with influence from Dutch ij~y (many early modern printers were from the Netherlands), so write > *wryte, though igh would probably survive as it is;
• With the above, I would expect pie, die to be spelled as pye, dye, which are archaic variants;
• I recollect seeing spelling aou, aow replacing ou, ow when pronounced /au/ in something (and I can't for the life of me remember what), and think this is actually a fairly sensible spelling — round, crown, bough > *raound, *craown, *baough;
• Otherwise ow, ough would probably represent the continuation of earlier /ou/, as in soul > sowl, as, bowl, throw;
• Remaining uses of "short o" to spell /ʌ/, one would think would be replaced with "short u" — done > *dunne, some > summe;
• And ou would probably be used in French words where pronounced /uː/, as in soup, and represent Middle English /uː/ where it did not become /au/, as in roum (an archaic spelling from Middle English); it might also replace oo as a spelling of /uː/ in other native words (moon > *moun), if oo were regularly then used to spell [ʊ].

English does have a vexatious vowel inventory, now doesn't it?

[Kuchigakatai]

author, authority should be either auctor, auctority after Latin or autor, autority after Old French and Middle English (a pronunciation of authority with /t/ survived into the 18th century, but was proscribed).

Interestingly, I have seen "author authoris" in 17th-18th century Latin... and I don't think it was in prints from England? I really wonder about the origin of this form of the word.

inchoate should be incohate (L. incohātus; in Medieval Latin apparently often inchoātus)

The Lewis & Short Latin-English and the Gaffiot Latin-French dictionaries don't seem to think inchoo is medieval though, but an alternative spelling. In fact, the Gaffiot dictionary uses inchoo as the main entry, incoho having a redirect to it.

Regarding my comment above on (h)ūmidus and (h)umerus, this is also true of both terms though... Maybe their h- forms are not medieval either.

Remaining uses of "short o" to spell /ʌ/, one would think would be replaced with "short u" — done > *dunne, some > summe

Minims really did a number on us here. Munth, wunderful, hunney... luve, shuvel, cuver...

[Kuchigakatai]

I just checked the online facsimile of the Vergilius Vaticanus (from ca. 400 AD), and I notice that in Aeneid book 6 line 252 there is in fact the line:



tum Stygio regi ? nocturnas inchoat aras·

(Not sure what that symbol between regi and nocturnas is, but it looks like an A, crossed, probably added by mistake.)

So inch- is definitely ancient at least, though we'd need a classical-era inscription (or fragment) to confirm it's classical...

[Richard W]

While reading the news on Putin's war, I was startled to see a neuter single adjective ліквідовано 'liquidated' (or, better, 'eliminated') plastered across a picture of the recently slain Major-General Vitali Gerasimov in a recent Ukrainian release. On digging into it, I discovered it was just the key word of the fuller, grammatically unexceptional (for Ukrainian) sentence Під Харковом ліквідовано генерал-майора російської армії, which word for word is 'Near Kharkiv [there was] slain a major general of the Russian army'. However, the construction seems odd for your average Indo-European language. The verb is expressed by the neuter singular short form of the past passive participle, and the patient is in the form for the object of a transitive verb. In this case, as befits a masculine animate patient, it is in the genitive case: an inanimate patient would be in the accusative case. The word order is the example sentence seems a bit odd, but I am sure this is just a matter of topicalisation.

Am I right to think that this passive construction is odd? I have seen refusals to accept some semantically passive constructions as passive because of the oddness of their construction, e.g. the Thai adversative passive. I have read that this Ukrainian construction always eliminates (as opposed to demoting) the agent.

[hwhatting]

Yes, this construction is parallel not to passive constructions, but to impersonal cosntructions like Russian likvidirovali "(they) liquidated". I remember reading about this kind of construction somewhere ages ago; I think the examples discussed were Polish, where that kind of construction also is possible. While the verb form is passive morphologically, I would rather file them under impersonal, because while the agent is not in the subject role, the object is not promoted to subject.

[Richard W]

I suppose a parallel would be the Latin impersonal passive, but that seems to be vanishingly rare with transitive verbs - one of the few examples is Ennius's vitam vivitur 'one lives a life'. A possible path would be to passivise a sentence with a cognate accusative (an idea I've taken from Maria Napoli). In Latin, the gender agreement in the verb is neuter. Then, if extending it to include a patient, the verb forms often disagree with the patient in gender, so it may make more sense to keep the patient in the object case.

[hwhatting]

I assume that in Slavic, the pathway rather is "wrong" extension from impersonal constructions without overt object (examples in Polish):
Zrobili (formally 3PL active ) "It has been done" (lit: "They did it", but "they" doesn't specify any concrete agents here) => Zrobili pracę (work-ACC) "The work has been done" (lit "They did the work").

Zrobiono (formally past passive participle, neuter nom. SG) "It has been done" => simple extension to Zrobiono pracę with "work" in the accusative, as above, instead of transformation to Zrobiona praca with "work" in the nominative and the participle agreeing in gender with "work", which, BTW, is also a totally valid construction.

I don't know if such simple extension is possible with the reflexive-base impersonal construction (zrobiło się => *? zrobiło się pracę); I don't think I have come across this, but only across the "transformed" construction zrobiła się praca, with "work" in the nominative and the verb agreeing in gender with "work".

[bradrn]

Some most fascinating poems I just found: the d’Antin collection, as edited and glossed in van Rooten’s Mots d’Heures: Gousses, Rames. These are clearly rich and meaningful texts, as amply demonstrated by just one of the annotated poems:

Et qui rit des curés d'Oc?¹
De Meuse raines,² houp! de cloques.³
De quelles loques ce turque coin.⁴
Et ne d'ânes ni rennes,
Écuries des curés d'Oc.⁵

L. d'A. van Rooten's illuminating notes:

1. Oc (or Languedoc), ancient region of France, with its capital at Toulouse. Its monks and curates were, it seems, a singularly humble and holy group. This little poem is a graceful tribute to their virtues.
2. Meuse, or Maas, river, 560 miles long, traversing France, Belgium, and the Netherland; Raines, old French word for frogs (from the L., ranae). Here is a beautiful example of Gothic imagery: He who laughs at the curés of Oc will have frogs leap at him from the Meuse river and . . .
3. . . . infect him with a scrofulous disease! This is particularly interesting when we consider the widespread superstition in America that frogs and toads cause warts.
4. "Turkish corners" were introduced into Western Europe by returning Crusaders, among other luxuries and refinements of Oriental living. Our good monks made a concession to the fashion, but N.B. their Turkish corner was made of rags! This affectation of interior decorating had a widespread revival in the U.S.A. at the turn of the century. Ah, the Tsar's bazaars' bizarre beaux-arts.
5. So strict were the monks that they didn't even indulge themselves in their arduous travels. No fancy mules nor reindeer in their stables. They just rode around on their plain French asses.

[Rounin Ryuuji]

5. So strict were the monks that they didn't even indulge themselves in their arduous travels. No fancy mules nor reindeer in their stables. They just rode around on their plain French asses.

If I'm not much mistaken, the French actually says "And neither donkeys nor reindeer/Do stable the curates of Oc."

[bradrn]

5. So strict were the monks that they didn't even indulge themselves in their arduous travels. No fancy mules nor reindeer in their stables. They just rode around on their plain French asses.

If I'm not much mistaken, the French actually says "And neither donkeys nor reindeer/Do stable the curates of Oc."

I don’t think you’re saying anything different to the annotation.

EDIT: Sorry, you are indeed correct. I missed the fact that ‘ass’ = ‘donkey’. (I don’t work with draught animals much.)