1. The WordReference Forums have moved to new forum software. (Details)

Irregular verbs: key verbs?

Discussion in 'Etymology, History of languages and Linguistics (EHL)' started by ThomasK, Nov 25, 2009.

  1. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Could one say that the key verbs in a language are very often irregular and v.v. ? If yes, why ?

    If not, how do you explain that the most common verbs are often iirregular in the Western European languages that I know ?
  2. Alxmrphi Senior Member

    Reykjavík, Ísland
    UK English
    Fascinating question!
    I'll be really interested to see what others' view is of this subject, but the same thought has often crossed my mind, especially considering the verb 'to be' in some languages.

    I don't really know anything about it but my speculation was as the key verbs are often the most common, they're more subject to change through regular usage, but I can't really back that up with any evidence.

    What examples can you draw from? I.e. what irregularities have you noticed (in what languages), I'm interested to see if they link up to any similar verbs in verbs I have noticed are irregular in other languages.
  3. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    I refer to Dutch and German, also to English: just have a look at the irregular verb list and you will see that the most basic words are there.

    I would say they do not change a lot because they are that often used, but that might not be quite true. At any rate I seem to get quite a lot of what I would call 'basic' verbs when looking at that list.
    I must add though: 'horen'/ 'hear' is quite regular with us, but see is not, which seems strange to me --- except if hear were some kind of causative, but that is not self-evident..;. I think there are fewer irregularities in Italian, more in French.
    By the way: they are all relatively short, which might betray their age as well, I think, if you understand what I mean...
  4. entangledbank

    entangledbank Senior Member

    English - South-East England
    Children learn irregular verb forms as separate words before they learn the rules. So they acquire the word 'ran' before they know how to make past tenses. Then they acquire the rule and go through a phase of saying 'runned', before finally working out where the irregular forms fit in the system.

    So I suppose the more common a word, the more likely it is they have learnt it early and thoroughly, so it's protected against replacement by analogical processes. Most irregularities are originally because of regular phonetic processes (e.g. Germanic umlaut such as man/men or PIE ablaut such as drive/drove and [Proto-Germanic?] was/were); these apply to all words, not just the common ones, but analogy obliterates the less common ones.

    There are also processes that do apply only to more common words: functional words phonetically change in ways lexical words don't (e.g. the English future marker gonna). This will certainly apply to some of the forms of 'be'.
  5. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Quite right, it seems to me. But any reason why 'hear' is less irregular ;-) than 'see' ?

    How come 'work' is not irregular (in any of the five languages I know - although in my dialect there is a very old work (wroeten, wrocht, ...) that was irregular). [Did/ do we start working only later on ? This does not make sense as such, but my feeling is that referring to something as 'work' is a cultural interpretation, non-existent in certain 'primitive' cultures]...
  6. entangledbank

    entangledbank Senior Member

    English - South-East England
    English also has the old past wrought; and English heard vacillated between regular and irregular quite recently: in the eighteenth century good speakers were divided how to pronounce it. Analogical levelling can affect even the most common words: it's just a statistical tendency that commonness matches irregularity.
  7. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    But isn't there a statistical tendency in both E. and Dutch for verbs to become regular? It is at least the case in D. : kloeg has become klaagde (complained), joeg jaagde (hunted, chased), etc. yet they are quite common, I'd say...

    Don't we perceive different trends ?
  8. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    I don't see a correlation between frequently used (key) verbs and irregularity an sich.
    I see a lot of basic verbs which are irregular (e.g. zijn en hebben), I see a lot of obselete and less frequently used verbs which also feature the list (werven, bieden), and I see a lot of verbs which I cannot call either "key verbs" nor otherwise, such as schrijven (to write), zwemmen (to swim), etc. which also feature the list

    So, my guess is that, for a verb to be irregular* (at least what contemporary European IE languages are concerned), it has to be used quite frequently, often, sometimes or hardly at all. Not that I would call this an explanation...
    What I am trying to say is that the extremes are easily to account for, it's the large group in the middle, neither key nor obselete nor infrequent, which pose a real problem...

    [* And let's not forget that most of the irregular (strong) verbs of today once were incredibly regular :)].

    Last edited: Nov 25, 2009
  9. berndf Moderator

    German (Germany)
    Ok, let's concentrate on Germanic.

    As Frank mentioned in his footnote, the term "irregular" verb is historically very misleading. Germanic strong verbs developed out of PIE conjugation patterns which were regular but strong, i.e. changing the root vowels rather that by suffixes only, as it is the case with Germanic weak verbs.

    The standard theory is that strong verbs were the original verbs inherited from PIE while the weak verb were later formations, probably using enclitic auxiliaries, e.g. he loved might have developed out of a form meaning he love-did*. The precise origin of the weak conjugation pattern is not known.

    This means, that strong verbs are the most basic ones because they are the most ancient ones.

    Again as Frank indicated, the original regularity of strong conjugations is today obscured. This is because the forms ceased to be productive a long time ago and different forms developed differently. The verb to see you mentioned is derived from PGerm. *sekhwan and different forms were contracted already differently in OE (ic sēo, hē sihþ, preterite plural sāwon). In its PGerm. original form, the verb follows the V-class conjugation (give-gave type).

    *The verb to love was derived from a noun (OE. lufu=love) and/or adjective (OHG liub=ModHG lieb) and is not an original PIE verb.
    Last edited: Nov 25, 2009
  10. Joannes Senior Member

    Belgian Dutch
    Yes, and shouldn't we therefore rather be talking about verbs like 'to be' and 'to have' (and not just Germanic), because there post #1 does have a point..

    (Tenzij je het effectief specifiek over sterke werkwoorden had, natuurlijk, ThomasK..)
  11. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Thanks, gentlemen. Please understand that we are just exploring a hypothesis ! ;-)

    Interesting consideration: '(ir)regular' is not purely denotative. However, the main point is whether that kind of conjugation could betray something. But Berndf's explanation seems to suggest that those verbs were the oldest.

    May I suggest we focus on the question whether the fact that 'work' and 'love' do not belong to that category whereas 'swim' does, shows that the hypothesis does not make sense.

    At any rate: my feeling is (was ???) that learners who do not know many of those 'strong' (...° verbs, miss out a lot if they want to communicate. (So, indeed, Joh., I did mean more than "be"' and "have"...) That was my starting point, in fact, but I did not mention it.;-)
  12. TitTornade

    TitTornade Senior Member

    In French, verbs are traditionnally classified in 3 groups :
    - the first group, the -er endings verbs (around 4000 words that are almost regular): it is the only group of verb that is “alife”: all the new verbs belong to this group. Most of them are seldom used.
    - the second group, the -ir ending verbs (around 300 verbs) : it is a group of regular verbs.
    - the third group, that contains mainly irregular verbs (about 370). The most common verbs belong to this group: they are the most irregular verbs!

    Of course, conjugations in French (and in roman languages) are much more complicated than in English (and than in Germanic languages). But, I think we can’t say that all the most regular verbs came from recent construction and that all the most irregular are from older construction; the patterns of conjugations often derived from latin.

    Thus, are the irregular verbs the same in Romance and Germanic languages? (If there are the same in each group… :p)
  13. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)

    I find this an interesting topic. But one thing is slightly unclear to me: what exactly do you mean by key verbs?
    So far, I have the impression that it is a very subjective, intuitive notion.

    I think we can agree that "to have" and "to be" are quite "crucial" verbs in the Western IE languages. But apart from those, what exactly is the criterion (or set of criteria) to decide whether or not a verb is a "key verb"?
    Irregularity (in itself?)? Definitely not.
    Frequency? I think I made a case in a previous post against frequency. And besides, if we'd take this into account, then we have to look at the frequency of verbs in different eras (which, by the way, would yield very limited results at best, useless results realistically).


  14. TitTornade

    TitTornade Senior Member

    Frank : I think you can speak French, right ?
    The 11 most common French verbs are : être, avoir, faire, dire, aller, voir, savoir, pouvoir, falloir, vouloir and venir (to be, to have, to do/make/get..., to say, to go, to see, to know, can/may/might/could..., to have to, to want, to come)

    Most of them are used as auxiliaries or as modal verbs : is it the meaning of "key verb" ?
  15. Erick404 Senior Member

    Rio de Janeiro
    Portuguese - Brazil
    I've read once that agglutinative languages tend to be much more regular. Quechua for example doesn't have a single irregular verb. This could bring more to think about...
  16. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    OK, as for 'key verbs' : I hope it is not too subjective, but I meant verbs that verbs that have a high frequency or maybe have a high frequency as an 'semantically underlying' verb.
    By the latter (non-English) I mean: 'to drag' refers to pulling, and that is a key movement, I think, that anyone should be able to describe when learning a language. I would like to find a list as a matter of fact of such verbs, but I think I could do it based on a list of the 5,000 most frequent words in Dutch.

    Another typical aspect of those KV to me is that they often lead to phrasal verbs or 'afgeleide werkwoorden'.
    Last edited: Nov 26, 2009
  17. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Interesting information as such, but I suggest we focus on a possible link between grammatical 'deviancy' (apparent) and frequency/ importance of verbs...
  18. sakvaka

    sakvaka Moderoitsija

    This is true. For example, Finnish has only two irregular verbs (to be and not), but there are 27 conjugation types. However, each conjugation has the same suffixes - only the word stems differ.
  19. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    I sometimes think so too, but alas ;-). I understand French fairly well, though.

    I made quite a similar list in my head too.

    Moderator's note:
    Let's stick to Western European languages in this thread, as suggested in the original post. Looking at how the thread developped, I think we have to interpret this as Indo-European languages, more specifically Germanic and Romance.
    Once again, this doesn't mean that the verbal system in Quechua or any other language doesn't deserve a thread on its own!


  20. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    So I just wish to remind you, if you don't mind: I have explained my hypothesis a little more in #16. Does it make sense to any of you ? The modal-verb issue is not what I am interested in, that is way too narrow to me now...
  21. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    I don't know, but post 16 is still rather vague. You have a hypothesis, you want to use a set of key verbs, but your criteria for compiling that list of key verbs is still too vague and hence the (lack of clear cut) criteria opens the gate widely for a confirmation bias.


  22. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    I am not sure whether the criteria are not clear cut. I mean: I am referring to verbs generally presented in the 'irregular verbs' list (not mine). And then I ask: can they in some respect be called 'key verbs'? Are they in some way more important than others - for example having a broader semantic scope (which might explain why they allow for prhasal verbs/...)? Do they perhaps describe certain 'basic' [human] movements?

    In fact if we can find another criterion explaining why those are 'irregular', fine with me. My hypothesis is simply: they seem like more important than a lot of others - but I guess I am biased somehow. So are some scientists. I am quite willing to accept that, but please prove the opposite (the hint at 'to swim' is not a bad one, but on the other hand, it is the basic human movement in the water...).
  23. CapnPrep Senior Member

    I would say that you were a bit hasty in your previous post:
    Yes, we can find frequent verbs that are regular and infrequent verbs that are irregular. But there can still be a correlation between frequency and irregularity. As a very general principle, linguists seem to believe that such a correlation exists. See the discussion of the "conservation effect" here, for example. And this makes sense from the point of view of language acquisition and language evolution: as others have already mentioned, children tend to regularize, and succeeding generations of speakers tend to regularize, unless there is sufficient pressure to maintain irregular forms. This pressure typically comes from frequent usage (but this is not the only possible motivation).

    Specifically for German, the facts are perhaps not so clear. For one thing — as a couple of people have already pointed out — strong verbs are not exactly irregular. For another thing, there are not only strong and weak verbs, but also "mixed" verbs, and there is a question of whether to count prefixed verbs separately, or group them together as instances of the same base verb. For example, this table says that 502 of the 1000 most frequent German verbs are strong verbs, but later in the text, the author explains that in fact, 150 of the 1258 Grundverben are strong verbs (and the rest are weak or mixed). This study doesn't specifically say that the strong verbs are the more frequent ones; the author just takes this for granted. I suspect that this result has already been confirmed somewhere else. It's not difficult to do, just time-consuming. One simply needs to look up the relative frequencies of all the verbs in the list, for example here (for the verb schreiben), and then do some quick statistics.

    Edit: Here's (the middle of) another discussion of this issue, from a diachronic perspective.
    Last edited: Nov 26, 2009
  24. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    Idem dito for "zinken" (irr.)? (It's my basic movement in the water.)

    Your example illustrates exactly what I mean. You almost decree that 'to swim' is a basic human movent in the water. The criterion here is "because I think so". If it would be a basic movement in the water, so what? What would basic movements have to do with the irregularity of the verb? Is graven (irr.) a basic movement in the soil and hence irregular? Houwen (irr.)a basic movement in rock? And hakken, oops, not irregular hence not basic?

    The example makes clear that the lack of fixed criteria let the hypothesis and your willingness to confirm that hypothesis, determine which verb is to be regarded as a 'key verb'.
    I can't wait for your explanation about how basic "(uit)pluizen", "schrijden", "krimpen" etc. are. And how "un"-basic "maken" is, and "werken", "betalen" etc. etc.


  25. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    You might be right. I have been starting from a more limited list, had left out the ones you mention (especially like 'pluizen', 'schrijden'), and had already admitted that the fact that 'werken', 'maken' in Dutch, might indeed be considered counterexamples. Mind you : 'kopen' is 'irregular' with us, 'betalen' is not.

    I still point out: did I decree anything? No, I tried to suggest a possible explanation. And oK, I had proposed a hypothesis - and I thought I had some reasons to do that.

    I would still like to point out that for example 'trekken' (pull) seems more basic to me than 'drag', cull, dislocate, drag, evolve, extract, gather, haul, heave, etc.; idem ditto with 'draw'. That was one of my starting points, and I really wonder whether that is ill-founded or whatever... Then, the next hypothesis was: maybe there are some things that one can do with one's hands, mouth, ears, etc. and which can be considered basic. Is that that unimaginable? And then, next, came the idea: aren't many of them in the list of irregular verbs? Is that a coincidence ?

    I know that some kind of wishful thinking is a hobby, but I welcome all kinds of comments...
    Last edited: Nov 26, 2009
  26. berndf Moderator

    German (Germany)
    I think it is a valid approach in science if you find an obvious correllation (and there is one here, however you define "basic") to start with fuzzy definitions and refine them as your understanding progresses. The only thing to be avoid are ad hoc alterations in order to "immunize" your pet-theory against criticism.

    So, I wouldn't want to require Thomas to provide a bullet-proof definition as a pre-requisite for discussion. And I didn't see him shifting definition in an "immunizing" way.

    I think the definition which comes with the standard theory, namely that strong verbs are almost always (I say almost because there are some cases of later formation of strong conjugations by analogy) inherited PIE verbs, gets us already most of the way towards a definition producing testable theories of high explanatory power. E.g. it explains why frequent and elementary weak verbs like to love and to work can exist based on verifiable data.

    One of the things left to be explained is the phenomenon of the replacement of strong by weak conjugations, e.g. 19th century German der Hund boll vs. today's der Hund bellte, and why certain verbs are (maybe?) more likely effected by this tendency than other. Frequence might play a role here but als intutiveness of the conjugation scheme, i.e. did er bol disappear because bellen is infrequently used in preterite? I don't know, I haven't gathered statistics. Or is er bellt-er boll an unituitive scheme? I don't know either; we would have to find testimony from a 19th century speaker.
  27. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    I need to ask: what definition do I need, and what is the link with love and work (which seem counterexamples) ? I thought of this (while not sleeping well):
    - 'groeien' (grow) might be a newer variant of the older 'wassen' (German 'wachsen') traces of which we still find in 'gewas', 'uitwas', 'volwassen' [though we only have 'groeien' in my dialect)
    - 'werken' (work) : 'wroeten' (still a common alternative in some West-Flemish dialects)
    - 'beminnen' (love): 'graag zien/ hebben' (to see with pleasure (?), to love to see, to be eager to see perhaps) ???
    So there were some strong verbs before, but then: why replace ? Or why drop them?

    Indeed, some that were strong before, have become weak, but then there is a trace in the perfect participle in Dutch (and German): -en. As in 'bakken' (bake), [boek], gebakken; 'wassen' (wash), [wies], gewassen. There might be some more. Why? Some kind of analogy ? The preterite/ imperfectum theory does not seem to have explanatory power as for 'bakken' and 'wassen', I think: we must certainly have used a present perfect of those verbs.

    As for the definition: are you suggesting, Berndf, that a definition of some kind should allow to de-fine, limit, ..., the 'basic verbs', in some semantic way ? (Sorry, I am at a loss here)
  28. berndf Moderator

    German (Germany)
    If we are talking about the origin of the dichotomy in PGerm of strong and weak verb, "inherited from PIE" seems to be the most useful definition of "basic". If we are talking about the loss of strong forms in modern Germanic languages, there might be different mechanisms at work and a theory explaining this phenomenon will need different terms.
  29. Meyer Wolfsheim Senior Member

    East Egg
    I have to disagree with you on one thing: Romance verbs are not more complicated that Germanic ones. Germanic verbs take internal modifications where as most romance language verbs only add suffixes and if there is an internal vowel modification, it is almost always predictable, like Spanish venir>viene or French tenir>tiens, because the natural phonology of the language accommodates the change, where as in English there is no phonology to make "give" to "gave" seem inherently apparent.

    What I am saying is there appear to be more "natural" changes in the Romance languages than in Germanic languages; Spanish a+el makes sense to become al and similarly de+el to del. Even the french à+le and de+le to au and du respectively can be "derived" and understood by a non-native. However, the same is not true for a Germanic strong verb.

    Despite being a native of English (a "Germanic" language), I find Romance language verbs much less complex than those in other Germanic languages, such as German where not only the conjugations are more complicated and less phonologically friendly, but the auxiliarly verbs as a replacement for inflecting for a mood/tense also are a burden.

    I think often the case is that there is some set of core verbs which can be found to be irregular throughout most Indo-European languages and generally it is possible to infer whether a verb will be irregular based on its stem/infinitive given your own knowledge of the regular conjugations.

    Strong verbs are not really irregular but what makes them so is that the whole umlaut/ablaut verb tense system is no longer functional; all new verbs in English always become weak verbs; and the fact that there are several conjugation groups within the strong verbs, thus misleading many to instantly classify them as being irregular.

    I remember when I was in Middle School we were quizzed on strong verbs (list past participle and preterite form of any given verb) and surprisingly most students (all natives of English) failed to get a mark above a B.
    Last edited: Nov 29, 2009
  30. TitTornade

    TitTornade Senior Member

    Of course, the question is not to know if the pattern of irregular conjugations is more difficult in Romance or in Germanic languages.
    What I was meaning is that for each strong verb in German (or in English), you have to know a maximum of three vowels (or group of letters) and you can conjugate the verb in all the tenses (German: lesen > er liest, ich las, ich habe gelesen: I learnt "lesen, liest, las, gelesen" as I learnt in English: "read, read, read"). :p
    (Meyer Wolfsheim: I'm curious to know how you learn irregular French verb.)

    In French, it is not always as simple as knowing 3 groups of letters.
    On the one hand, you have to know the many different endings and on the other hand, to know the different bases: some verbs (the regular ones) have one base and other verbs can have up to 6 different bases. And knowing the infinitive can’t always help knowing how many bases has a verb!

    Let’s rapidly compare (colours correspond to the bases, the endings must be ignored):
    Infinitive: voir (to see), avoir (to have) and savoir (to know)
    - je vois, j’ai, je sais (I see, I have, I know)
    - nous voyons, nous avons, nous savons (we see…)
    - j’ai vu, j’ai eu, j’ai su (I’ve seen…)
    - je voyais, j’avais, je savais (I was seeing…)
    - je vis, j’eus, je sus (I saw…)
    - je verrai, j’aurai, je saurai (I will see…)
    - voyant, ayant, sachant (seeing…)

    Voir and savoir have 5 bases : voi-, voy-, vu-, vi-, verr- / sai-, sav-, su-, saur-, sach-; but they are different and not used with the same tense. Avoir has many bases (not easy to count, it often merged with the ending). :eek:

    Of course, I think it would be crazy to create such new verbs with more than one basis, the endings being normally well known by everybody ;) Nobody can't imagine it!

    Another information : we have in French some irregular verbs such has "ouïr" (to hear) or "choir" (to fall) that have too complicated conjugations (and missing tenses...). These verbs can be considered as key verbs, but are almost not used at all now. They are replaced by more regular verbs : "entendre" (to hear) and "tomber" (to fall).
    Is it a phenomenon observed in Germanic languages? Did some strong verbs become weak verbs (in losing the changings in vowels)? Have some of them been replaced by weak verbs?
    Last edited: Nov 30, 2009
  31. palomnik Senior Member

    Frank: After re-reading the original post by ThomasK, I'm not sure that it was his intent to limit this to IE languages. And actually, I'm not sure it's a good idea to do that; there's no reason that a phenomenon like this would only exist in Europe - and in fact, it doesn't. Any comments?
  32. Meyer Wolfsheim Senior Member

    East Egg
    That perhaps maybe, but you cannot deny the fact that in general Romance verb conjugations are much more phonologically friendly than their Germanic counterparts. And one of the bases you listed is in fact no longer in use (the passe simple) in spoken form and you forgot to list the imperfect subjunctive tense (another literary form), "j'eusse", etc.

    - je vois, j’ai, je sais (I see, I have, I know) All irregular but savoir/avoir have a similar pattern. Analogically, Spanish "he" and "se" sound exactly the same.
    - nous voyons, nous avons, nous savons (we see…) All regular.
    - j’ai vu, j’ai eu, j’ai su (I’ve seen…) There are not nearly as many irregular past participle forms as exist in Germanic languages.
    - je voyais, j’avais, je savais (I was seeing…) Imperfect is an incredibly regular tense (only one irregular imperfect etre?)
    - je vis, j’eus, je sus (I saw…) Most of the time the passe simple uses the past participle as a base, and the "je vis" mimics the vowel sound in Spanish "ver" to "yo vi."
    - je verrai, j’aurai, je saurai (I will see…) There are only a handful of irregular future/conditional stems and once you know the core set, you can predict if a future stem will be different. For example, because venir>viendr-, it is acceptable to assume that tenir>tiendr- (the same exact pattern is in Spanish, venir>vendr-, tener>tendr-). "Verrai" is analogical to Spanish "vere."
    - voyant, ayant, sachant (seeing…) And of course you also have etant; the first present participle accords with the general rule of formation, while the second have a subjunctive base, irregular indeed, but again only limited to a handful of verbs.

    With a knowledge of Spanish verbs it is quite simple to understand French conjugations. However, with a native knowledge of English verbs, it is still far more difficult to understand German verb conjugations, despite their "simple" bases.

    While strong verbs don't seem to be very frequent in Romance languages, through after-analysis, one could conjecture that Spanish "eres" to "eras" could be a type of internal vowel modification to form a new tense.
  33. CapnPrep Senior Member

    Meyer, you and TitTornade evidently both know something about Romance and Germanic conjugations, in particular you agree on how to conjugate these French and German verbs. But you seem to "feel" differently about how "friendly" or how much of a "burden" they are, and which system is "much less complex" or "far more difficult to understand". You can go back and forth about this all day long, and never reach an agreement, because these feelings cannot be measured and compared in any objective manner. More importantly, knowing which of you is right and which is wrong will not help answer the original question of this thread.

    All of the French, Spanish, and German verbs you two have mentioned count as irregular in the context of this discussion, because they are not completely regular. Romance and Germanic languages both have a significant number of them, and the question is whether there is any meaningful way to characterize/predict the kinds of verbs that we find in this irregular group.
  34. sokol

    sokol Senior Member

    Vienna, Austria; raised in Upper Austria
    Austrian (as opposed to Australian)
    Much has been said already; here my comments as a late-comer to this thread:

    It is indeed the case that many common words are more irregular than less common ones; if you take Indoeuropean languages this is easily demonstrated with the copula - the forms of "to be" in IE languages: here we find several roots used (see the corresponding Wiki article on IE copula, and don't let yourself be irritated by the "laryngals" = h1, h2, h3 - you will easily recognise your native language copula roots if it is an IE one :) also they're listed below too).

    [I really recommend reading that Wiki article; it may be clear(er) to you afterwards, ThomasK, that probably (!) the IE copula "originally" was perfectly regular but this only is hypothetical as we can only recover the numerous roots for the copula, and how they might have come to be used as copula in the first place. We only know that at a specific point in time - before the IE branches spread over the Eurasian continent and lost contact to each other - several roots were shared by most of them, and some by all.]

    The principle is simple: what is repeated over and over again in everyday language usually is more resistant to change (but change there was, even with the IE copula - as shows the Wiki article).

    However, if you are searching for some logic why "hear" is less irregular than "see" you probably won't find any (and it would be splitting hairs to try and argue with statistical frequency of those words): repetition preventing change or even facilitating new irregularities (as it seems was the case with IE copula, see the Wiki article) only works as an argument for the most frequent words.

    The rest needs more complex explanations (no good example of an irregular root comes to my mind - it should be one used in more than a few IE languages to illustrate the point).
    Statistics definitely wouldn't be the solution for explaining them. :)

    Also please note that a few very common German verbs are perfectly (or for the most part) regular, for example the very frequent modal verbs (present and preterite: regular: müssen, wollen, sollen; irregular but only with umlaut, regular ending: dürfen, durfte; können, konnte; mögen, mochte; irregular - ablaut, no ending in preterite: lassen, liess).

    There is indeed a correlation between frequency and irregularity as CapnPrep suggested above - but I am sure that if somebody would take the pains to try and establish a theory based upon this correlation this theory would be doomed to fail.

    For some reason some of the most frequent verbs in IE languages indeed are very irregular, but it is not so easy as to make a general rule out of it - be it about IE verbs or frequent verbs, or words, in other languages.

    And Palomnik, I think you have a point: probably it would be helpful to compare with non-IE, non-European languages, and ThomasK didn't explicitly exclude them even though discussion so far mainly was about IE languages.

    However, I don't think the original question will be helped with listing plenty of paradigms (be it of IE or non-IE langauges): this just would make this thread an endless list which nobody would be willing to follow anymore (or I certainly wouldn't :D).
    So let's please stick to a few examples, let's try and keep it simple, that's my suggestion. ;)
    That'll be challenge enough to those of us who know nothing about more exotic languages. :)
    Last edited: Nov 30, 2009
  35. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Thanks, Sokol, very interesting indeed. I may be allowed to point out that my main focus was not on the irregularity as such - it reminds me of the argument on the sex of the angels ;-).

    I am interested in whether those verbs that are now 'strong', 'irregular' verbs (i.e., that students of a foreign language cannot deduce from the rules), that particular group, can also considered 'basic' from a pragmatic-semantic point of view.

    Frank06 (or is it 007) did made a fairly strong case against my assumption, I must agree, by referring to special examples like shrinking, investigating (uit-pluizen) in Dutch, which do not seem that important. Yet, the funny thing is that most people do seem to feel the same way as I do, and refer to frequency (which also implies 'communicative importance [importance ???]', I think, at least to some extent) in order to explain irregularity.

    Yet, it remains funny that 'work' is generally regular, but that may be due to culture (we did not realize we worked for some centuries ????), and 'make' is very often also regular, which is the strongest counterargument, I thiink.
    Last edited: Dec 2, 2009
  36. berndf Moderator

    German (Germany)
    To repeat my point in a nut-shell: "No, but":
    - "No", because there are simply too many counter examples.
    - "But", because there is nevertheless a significant correllation. And the explanation of this correlation (in Germanic languages) is that strong verbs are older and older verbs are often but not always the more basic ones.
  37. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Thanks. Just wondering: are they ways out? (Not because I want to prove my point, just wondering !).

    I remember for example the reference to the French older words for hearing, the older words in Dutch for working. How come some very 'basic' words are sometimes replaced ? How can they, if they are that important ? (I do not remember Guy Deutscher referring to that in The Unfolding of Language)
  38. sokol

    sokol Senior Member

    Vienna, Austria; raised in Upper Austria
    Austrian (as opposed to Australian)
    Berndf has summarised the gist on which - it seems - already a few of us agree upon here; but I'm a bit puzzled about this sentence, what do you mean by a "way out"?
  39. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    A way out... of the falsification of the hypothesis (as now it is "no but"). I thought of other arguments sustaining the hypothesis that they somehow belong together as 'basic':

    - a lot of them allow for phrasal verbs in English, samengestelde/ derived verbs in Dutch and German (think of all the 'stellen' derivations, both in German and Dutch) [by the way: why is 'stellen' no longer common in Dutch, and is it replaced by 'zetten' ?)
    - quite some seem to refer to parts of the body, limbs (hands, feet, mouth, the senses, ..;)

    Hope I am not stupidly insisting: I think I am still exploring this group of verbs...
  40. berndf Moderator

    German (Germany)
    One should be very cautious with this line of thinking because it usually leads to tautologies where the definition "basic" is twisted until it means "being a strong verb".

    But one thing to bear in mind is that "basic" is always dependent on the cultural, technological and environmental conditions. E.g., for us to swim might be a "basic" verb. For a people living in a desert the action might be something extremely exotic. Or a Stone Age civilization will probably not have a word for to forge.
    OE wyrcan had a past tense ablaut, he worhte, yet it is a weak verb. This is an additional complication.
    There are many possible reasons why a root can be lost. The precise reason will often remain unknown because some of these changes happened long before the oldest attested Germanic language. But for a few we have reflexes of the old verbs in OE and/or OHG which gives us a clue. E.g. the verbs
    beon = to become and wesan = to remain, to rest, to exist were lost when they merged with the is, sind root (cognate of Latin sum) to form the OE verb beon-wesan > ModE to be. Similarly, Latin stare = to stand is lost in modern French because it merged with esse = to be: stare>estare>ester>être, esse>essere>estere>estre>être.

    Another possible reason for replacement of a word it that it gets "competition" from imported or newly formed words and the "old" one starts to shift its meaning, like English haven getting competition from harbour and port and is eventually confined to a figurative meaning.
  41. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    I am about 2.5 years older ni the meantime, but I just happened to read something interesting in this connection:

    So: strong/ irreg. verbs can survive due to a frequent use. Verbs that are often used apparently have a meaning that is much needed.
  42. sanne78 Senior Member

    At the end of the article is this quote:

    Geschiedenis: sterke werkwoorden zijn oude werkwoorden
    ...Sterke werkwoorden komen uit een tijd lang geleden dat het Nederlands - of een voorloper daarvan - anders werkte en werkwoorden in de verleden of voltooide tijd een andere klank kregen.
    ‘Onregelmatige werkwoorden zijn vaker primaire acties, lichaamsbewegingen en dergelijke. Zolang mensen praten over zichzelf hebben ze deze begrippen nodig gehad. En zo overleven de onregelmatige vormen...

    I fully agree with this. I often tell my students that irregular verbs are the oldest / the most essential verbs. Real "key verbs"!
    Among others: to be, to have, to give, to go, to bring, to come, to do, to say, to sleep, to sit,..

    I have no knowledge of the history of languages, but it seems obvious to me.
  43. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    And what do you tell about irregular, strong verbs which aren't really key words? bieden, delven, bergen, blijken, (be)driegen, glimmen, kerven, krimpen (geen 60° was in die tijd), schrijden, ...
  44. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Aaarrhhhh, here is the 'old' argument again! Well, I can't deny that, Frank, I mainly mentioned the article because it seems to corroborate my hypothesis whereas I cannot but admit that delven, bergen, glimmen, kerven, krimpen (geen 60° was in die tijd), schrijden, ... do not seem to be indispensable. I do not think that is true for bieden, (b)lijken, (be)driegen; I think those are quite common and useful (not easily avoidable), just like bergen, referring to hiding (ver-) seems more useful already, even now.

    Just one hypothesis :
    - delven, referring to digging into the earth, might have been more important then
    - glimmen: no idea - kind of synonym of schijnen (to shine) ?
    - kerven: might that be the first verb of cutting ???
    - krimpen : a form of getting smaller ?
    - schrijden: the alternative for going when slowly ?

    It might be interesting to find out whether these or similar verbs in other languages are strong as well. In English it seems to work except for kerven (carving).

    But it is only a hypothesis.
  45. Istriano

    Istriano Senior Member

    I would say voltar (''to become'') is a key verb in Brazilian Portuguese, but it's 100% regular ;)
    Falar (''to speak'' but also ''to tell'' and ''to say'') is also regular.

    Portuguese has fewer irregular verbs that Spanish and Italian. In Spanish, even gerunds are many
    times irregular (yendo) or semi-irregular (durmiendo)...
  46. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    I do not mean that it is a general rule, but how about the ones that are irregular, Istriano? Are they important, teh important ones, do you think?
  47. sanne78 Senior Member

    Just a few things:

    1° I think that "to be" and "to have" are the most frequent verbs in most (European) languages; they are as well the most irregular.

    2° English is rather peculiar, as you can 'easily' distinguish the 'old' verbs from the 'new' verbs. The newer verbs are verbs related to modern concepts (electricity, computers, cars) and/or are of French/Latin origin.
    These are all regular.

    The older verbs are of Germanic (and ??) origin, are related to basic concepts (human feelings /activities) and can be irregular.
    Apparently verbs have a tendency (only in English??) to become regular: to dream-dreamed (instead of dreamt).

    Of course there are loads of exceptions, but to me this is the general rule.

    PS: Criticism of this point of view is welcome, because, as I said, this is just how I see it, and I have no in depth knowledge of the history of the English language (or any other language for that matter ;))

    PPS: I notice that for French people learning English the 'modern' verbs are fairly easy to learn, because they are of French origin (to continue, to insist, to invite, to finish), whereas the key/irregular verbs are much more difficult.
  48. Frank06

    Frank06 Senior Member

    Nederlands / Dutch (Belgium)
    If we would make a very black and white classification of contemporary Dutch verbs, using the axis irregular (or rather, strong) verbs and key verbs, then we get 4 (very artificial) groups.

    1. irregular + KV
    - verbs for to be, to have in quite a few IE languages
    - verbs which young native speakers learn after a certain age.
    I guess we all have heard dialogues as:
    Kid: Ik loopte (reg.)
    Parent: Ik LIEP (irr.)
    Kid: (slightly annoyed) Ja, ik loopte (reg.)
    But most native speakers wouldn't make that kind of mistakes anymore after a while.

    2. irregular + non-KV
    To me, this is the most interesting group, but quite a diverse one:
    (a) it contains irregular verbs
    - which are commonly mixed up, messed up if we'd use prescriptive standards (e.g. slagen/slaan, jagen, ...);
    - which have alternative forms, accepted by the global speech community, often codified in authorative dictionaries.
    (For example KV(?) zeggen, zei/zegde, gezegd (obs. zeide being an interesting form))
    Less clear examples: one can find instances of (incorrect) "stijgde" in online texts, but those forms hardly yield strong reactions. Try to post a text with "loopte", and the reactions are quite aggressive. One can hardly find an instance of "valde" (correct "viel").

    (b) On the other end of the scale, we have irregular verbs which are hardly used anymore in contemporary Dutch, which clearly belong to the written or more formal registers and which often seem to serve as a kind of peacock's tail.

    3. regular + KV
    praten, werken, betalen, ...

    4. regular + non-KV
    All the rest :)

    If I understand well, Thomas is solely talking about the 1st group, while, at least in my opinion, expanding or limiting the possible meaning of KV.

    Here it becomes tricky. Thomas doesn't seem to like a historical approach ("Aaarrhhhh, here is the 'old' argument again! "), but nevertheless, in his "pragmatic-semantic" approach, he smuggles in historical aspects, ad lib. on very subjective grounds, viz. his "semantic-pragmatic" p.o.v.

    Which makes me agree with Bernd's remark:
    I have less problems with your assumption, which could apply to class 1. I have more problems with your general approach, your vague terminology, and, excuse me, your fancy hyphenated words.
    I just tried to expand the little tiny hole in the wall, which could lead to tunnelvision, and indicate that it is rather pointless to talk about irregular strong KV if one isn't particularly willing to consider or doesn't want to take strong non-KV and regular KV into account (my artificial groups 2 and 3).
  49. ThomasK Senior Member

    (near) Kortrijk, Belgium
    Belgium, Dutch
    Let's start with your problems: I have no problem recognizing that I venture upon some hypotheses, trying to combine all kinds of information starting from some kind of intuition. And OK, I use those hyphenated words in order to explain that I think that my theory is somewhere in-between, because I think that explanation borders on semantics, maybe on pragmatics (meaning defined by context, I mean). The old problem was not a simple historic problem, in my view, but your reference to those verbs, delven, etc. which I recognize, and which I tried (!) to deal with in #44 --- because I recognized it as a valid argument.

    I don't understand why you think I don't want to include 2. I am inclined to consider slaan, maybe jagen, as belonging to the KV group indeed; I mean: they seem fairly important in every-day conversation, etc.

    Of course 'basic' is a fairly general word. But I am using it until I can maybe show that those verbs are somehow 'primary' (like with colours), wiht respect to others. And I love exchanges that force me to sharpen the way of formulating things, and grant me a chance to do some freewheeling (and maybe never 'land').

    What I happened to come across did seem interesting to me and seemed like some corroboration of my hypothesis.
    Last edited: Oct 12, 2010
  50. Istriano

    Istriano Senior Member

    Even primarity of colors is relative :)

    (Wikipedia) :D
    Dat is echt goed ;)

Share This Page