Progressive prescriptivism?

« previous post | next post »

I'm puzzled. The reason is that I've just read Merja Kytö, "Be/have + past participle: The choice of the auxiliary with intrasitives from Late Middle to Modern English", pp. 17-86 in Matti Rissanen et al., Eds., English in Transition: Corpus-based Studies in Linguistic Variation and and Genre Styles, 1997.

The content of Kytö's chapter doesn't puzzle me — it explains very clearly how English changed from be to have as the marker of perfect aspect in intransitive verbs. This change is easy to see in bible translations, where for example in 1 Samuel 26:20, the King James Version of 1611 gives "the king of Israel is come out to seek a flea", where the 1978 New International Version of 1978 gives "[t]he king of Israel has come out to look for a flea".

And the timeline is also pretty clear. Based on tracking the use of be/have + past participle in a corpus of about 2.7 million words spanning the period from 1350 to 1990, Kytö demonstrates that "in the late Middle English period, the use of have increases gradually, gains in momentum in the late 1700s and supersedes the use of be in the early 1800s".

What puzzles me is why this process seems to have escaped the censure of prescriptive grammarians. Here's a change that "[gained] in momentum in the late 1700s", just when the likes of Robert Lowth and Lindley Murray were in bloom. Did anyone stand up against the rising tide of have for marking the perfect in intransitives? If so, their delaying action was ineffective and quickly forgotten.

I don't know much about the history of prescriptivist ideology, but I think that I may be projecting the ideology of modern preservationist prescriptivism inappropriately into the past. In modern times, it's common for would-be linguistic authorities to pretend that they're holding the line against degenerative change, even when they get the history backwards. But John Dryden initiated the condemnation of sentence-final prepositions on the grounds "that the language, wit, and conversation of our age, are improved and refined above the last". Perhaps this style of progressive prescriptivism influenced some later grammarians to see the be/have change as a step forward into a better future, rather than a symptom of how far kids these days have fallen.

Anyhow, the be/have history is an interesting one. Here's how Kytö's chapter starts:

In many languages the choice of the auxiliaries be and have (and their equivalents) has been subject to variation in perfective constructions with intransitives indicating "transition" or "change" (cf. they are arrived versus they have arrived). In some Germanic and Romance languages the development has resulted in the generalization of one construction (e.g. the almost total dominance of "be" in Present-day Danish and "have" in Spanish and Portuguese); in others both variant forms occur in certain grammatically, stylistically or regionally restricted contexts (as, e.g., in Present-day Swedish, German, French and Italian).This cross-linguistic variation reflects the differences in the systemic realizations of the distinction between state (favouring be) and action/process (favouring have). Over the successive stages of development, various linguistic and extra-linguistic factors have influenced the choice of one form or the other.

In the Old English period the be/have + past participle construction denoted "state" in instransitive and transitive uses (cf. hie wæron gecumene; hie hæfdon hine gebundenne). The past participle originally functioned as an adjective and was sometimes inflected (Mitchell 1985:1, §709). However, the grammatical concord was gradually lost and the past participle was placed immediately after the auxiliary. Have originally occurred with transitive verbs only, but early on came to be used with intranstives, too. In Early Middle English be prevailed with mutative verbs, but have started gaining ground slowing in uses with the emphasis on "action" and the notion of perfectivity (Rydén-Brorström 1987: 17-18). Signs of the rise of have are apparent from around the 1400s on (for a recent corpus-based study, see Elsness 1989: 100; 1991: 276-283).

Kytö summarizes some ideas of David Denison's about what happened when, and why:

The development of be and have perfects show features attributed to the process of grammaticalization, but scholars disagree about the exact chronology and nature of the process. A recent discussion on the topic can be found in [David] Denison [English historical syntax: Verbal constructions] 1993: 340-368 … With have perfect the relevant factors have been the loss of inflections in the participle, the word order, and certain VP types (according to the semantics of OE habban and the valency of the main verb). The process of grammaticalization would have reached a stage of fulfilment "when the have perfect became available for any lexical verb which did not conjugate with be", i.e. when it became an auxiliary verb, the suggested (but debatable) date for this being the late Old English period (Denison 1993: 352). The other possible stages of grammaticalization include the point when the construction became a tense equivalent (probably in late Old English); when it had developed its present-day meaning and superseded be (probably in the seventeenth century); when it became used with all non-auxiliary verbs (in late Modern English) (Denison 1993: 352). There is, similarly, disagreement over the grammaticalization process of the be perfect (for a summary, see Denison 1993: 360-361); the factors regarded as having influenced the process include, e.g., the increasing use of be as the auxiliary of the passive and the relatively light functional load of have, the possibility of neutralizating the present tense third-person singular forms into the clitic 's, and the prescriptions of normative grammarians.

Over the two centuries since the change became essentially complete, several sources of variation — especially archaism and ambiguity with quasi-adjectival constructions — continue to generate examples that might be taken to mark the perfect with forms of be. Thus Wilfred Campbell in 1905 wrote a sentence with two apparent be-perfects ("is vanished" and "is gone"), one have-perfect ("has passed), and one be + past-participle-adjective ("is done") in parallel:

17 Where the woe that wrecked me is vanished,
18 And the pride that stayed me is gone:
19 And only the feeling of eventime,
20 When the toil of the world is done:—

21 O, Master of being and slumber,
22 When the pageant and pæan have passed;
23 Take me where thy great silence
24 Is vaster than all that is vast.

And a search of Google News turns up these current examples:

Return to the oven for 45-50 minutes, until the centre is risen and set.
We have to depend on well water that is become so orange because of all of the growth that I can no longer wash clothes at home.
He added, "Globalization is come to be known as an international avenue for profiteers to exploit workers wherever they can."
"As far as I am concerned, she is dead. I have told my children their Nana is gone to heaven," Fiona Porter said.
Most of the water is gone to the Mississippi (River) through the Embarras River, but Lawrenceville is quite similar to Montpelier with the river.
But that option is gone to New Orleans now.
An enraged and despairing Sita jumps into Ravana’s funeral pyre and dies at the end of the play, even as Rama moans “My punishment is come upon me now. My dreaded punishment is come “.
Now that they've been hit hard by losses on their huge mortgage exposures, that backing is become more explicit.

Rama's moan is probably meant to be quasi-biblical language, while "until the centre is risen and set" could be interpreted along the same lines as "until the centre is high and firm".

But whatever the source, examples continue to occur that could be interpreted as be-perfects, and it's an interesting issue in linguistic dynamics that these don't apparently generate any tendency towards a reversal of the historical process.



  1. language hat said,

    July 25, 2008 @ 9:20 am

    I don't think "is gone" can be lumped in here; it's a standard expression (you wouldn't say "the pride that stayed me has gone").

  2. Pekka said,

    July 25, 2008 @ 9:47 am

    It is possible that at least one prescriptive grammarian took notice. This distinction is mentioned in a note under the entry come in Webster's Revised Unabridged Dictionary (1913 + 1828). A link.

    Am come, is come, etc., are frequently used instead of
    have come, has come, etc., esp. in poetry. The verb to
    be gives a clearer adjectival significance to the
    participle as expressing a state or condition of the
    subject, while the auxiliary have expresses simply the
    completion of the action signified by the verb.

  3. Mark Liberman said,

    July 25, 2008 @ 9:52 am

    Language Hat: … you wouldn't say "the pride that stayed me has gone"

    I might. And LION gives 2,419 examples where poets have (or at least have used the word sequence "has gone"). Arguably these mean "has departed" rather than "is no longer present", but that distinction is hard to define extensionally.

    Gone also participates in the historical change along with other past participles: thus the KJV has 1 Samuel 14:17 "Then said Saul unto the people that were with him, Number now, and see who is gone from us.", while the New American Standard Bible has "Saul said to the people who were with him, 'Number now and see who has gone from us.'"

    It's true that gone is especially likely to be used with be in contemporary English, and probably the right way to describe this is to say that it's sometimes an adjective. But equivocation between be + past participle as a verb form and as copula + adjective has been part of the history of this process from the beginning, as I understand it.

    And there are other particular forms that retain a fairly high frequency of uses with be, like "come" and "become", where the adjective theory is less helpful.

  4. Mark Liberman said,

    July 25, 2008 @ 9:58 am

    @Pekka: The variation represented by "is come" vs. "has come" has often been noticed and described. What I haven't seen is any attempt to prescribe resistance to the overall trend towards substituting have for be in intransitive perfects.

  5. Arnold Zwicky said,

    July 25, 2008 @ 10:16 am

    Mark Liberman: "It's true that gone is especially likely to be used with be, and probably the right way to describe this is to say that it's sometimes an adjective."

    That's in fact what some dictionaries say: NOAD2 and AHD4 both have entries for the adjective "gone", separate from their entries for the verb "go" (with past participle "gone").

    Compare the adjective "past", which is etymologically "passed".

    Of course, these adjectives have coexisted with past participles in perfect constructions (with both "be" and "have") for a long time.

  6. Chud said,

    July 25, 2008 @ 10:52 am

    From the Book of Common Prayer (1979 US, and, it seems, other BCPs, but this was the one that was drilled into my brain):

    Therefore we proclaim the mystery of faith:
    Christ has died.
    Christ is risen.
    Christ will come again.

    Clearly the writers see a difference between "is risen" and "has risen".

  7. Karl Hagen said,

    July 25, 2008 @ 10:59 am

    The old grammarians didn't analyze these as perfects but as "neuter passives". Here is Goold Brown's explanation:

    'A few active-intransitive verbs, that signify mere motion, change of place, or change of condition, may be put into this form [the passive], with a neuter signification; making not passive but neuter verbs, which
    express nothing more than the state which results from the change: as, "I am come."–"She is gone."–"He is risen."–"They are fallen." These are what Dr. Johnson and some others call " neuter passives;" a name which never was very proper, and for which we have no frequent use.'

    The usually arch-conservative Brown recognizes the fact that these forms have become obsolete, saying they "may now be considered errors of conjugation, or perhaps of syntax."

    So part of the reason for the lack of complaint may be that the perfect made sense to them but the "neuter passive" was an oddity, and so a regularization was preferred on theoretical grounds.

  8. John Cowan said,

    July 25, 2008 @ 11:23 am

    A quick look at the KJV (in modern orthography) suggests that it represents a very late stage in this change, where only "is gone", "is come", and "is become" survive as BE-perfects in full vigor. I searched for the regular expression /\bis [a-z]+ed/ in an attempt to find BE-perfects of regular verbs, and all the ones I looked at were at least arguably adjectival. Some of these are ambiguous, particularly the ones that have a PP after the adjective/participle, like Deut. 5:30 And when she is departed out of his house, where a modern writer surely would write has departed; but there are many more like Jer. 5:30 A wonderful and horrible thing is committed in the land, where modern English requires has been committed.

  9. goofy said,

    July 25, 2008 @ 1:12 pm

    is committed is passive, so I'm not sure it counts.

  10. Theodore said,

    July 25, 2008 @ 1:31 pm

    All of the Google News examples given are in the third person singular and could be replaced with the clitic 's, where they would become ambiguous with regards to be/have.

  11. Bloix said,

    July 25, 2008 @ 3:19 pm

    "that backing is become more explicit."

    This is so unidiomatic that it would seem to be a proofreading error- it should be either "has become" or "is becoming." "Is become" can't be possibly have beeen used in this sort of business context.

    I'm reminded of the verse of the Bhagavad Gita that Robert J. Oppenheimer thought of at the test of the first atomic bomb:

    I am become Death, the shatterer of worlds.

  12. Evan said,

    July 25, 2008 @ 8:50 pm

    As noted, gone, come, and become are high-frequency examples. In fact every Google News example that was given is one of these three, except the first ("is risen"), which is probably adjectival. This makes me think that modern usage of be is just idiomatic and restricted to these few participles. I'm no master of dialects but if I ever read "I am travelled to Europe" I think I would do a double take.

    I would guess the reason modern prescriptivists have not chosen one form over the other is that be is seen as archaic ("old" being "correct"), but have is so wildly favored at the moment that proscribing it would be impossible.

  13. Theo Bromine said,

    July 25, 2008 @ 9:42 pm

    One construction that comes to mind is "I am finished" vs "I have finished". In my dialect (Southern Ontario, Canada), they are almost interchangeable, but I think "am" is probably used more than "have". This was a source of confusion for my anglophone kids learning French: "J'ai fini" means "I have finished", ie completed the task, but "Je suis fini" means "I am finished", in the sense of being dead!

  14. Mark Liberman said,

    July 25, 2008 @ 9:42 pm

    Evan: I would guess the reason modern prescriptivists have not chosen one form over the other is that … have is so wildly favored at the moment that proscribing it would be impossible.

    Yes, it would obviously be impossible to try to turn the clock back now. What interests me is that people like Bishop Lowth, in the middle of the 18th century, didn't try to intervene in the process.

    Perhaps in those days, preservation of older forms in the face of change was not seen as a virtue. In fact, Lowth's criticisms of "false syntax" seem to have been framed in a spirit of reform.

    This may be a commonplace observation — assuming that it's even true — but I'm not an expert in this history, and it's a new idea to me.

  15. Rick S said,

    July 25, 2008 @ 9:43 pm

    @Mark Liberman: You counted When the toil of the world is done as a be perfect, but I see it as copula+adjective. If it were a be perfect, shouldn't it alternate with have perfect? But *When the toil of the world has done (with intransitive do) sounds ungrammatical—or at least obsolete—to me.

    As for why this change hasn't been condemned by prescriptivists, didn't Denison say that it has (Kytö's summarization: "the factors regarded as having influenced the process [of grammaticalization of be perfect] include…the prescriptions of normative grammarians")? I couldn't find English Historical Syntax online, but with 400+ examples cited (according to one reviewer) it seems likely some were quoted therein.

  16. Justin L said,

    July 26, 2008 @ 1:16 am

    Could the prestige of French or other Romance languages contributed? When I first started learning Spanish, I was struck by the (admittedly superficial) resemblance between the Romance derivatives of "habere" and "to have".
    Looking at the history of Romance, Menendez Pidal gives the 16th century for the generalization of "haber" in Spanish and the extension of "avoir" in French seems to predate standardization in the 17th century. 18th century English grammarians, looking to other languages–particularly French–as examples would have found that they generally used "have" perfects even with intransitives. Could this have contributed to their reluctance to condemn this development?

  17. Mark Liberman said,

    July 26, 2008 @ 5:31 am

    Rick S: You counted When the toil of the world is done as a be perfect, but I see it as copula+adjective.

    You're right, I was careless on this point and I've changed the text of the body of the post appropriately.

    … didn't Denison say that… "the factors regarded as having influenced the process [of grammaticalization of be perfect] include…the prescriptions of normative grammarians")?

    Yes, but I believe that this summary refers to the effects of normative grammarians in facilitating (rather than holding back) the shift from be to have.

    Karl Hagen's citation of Goold Brown on this point is an example, I think.

  18. Steve Harris said,

    July 26, 2008 @ 1:36 pm

    Is there possible connection with the German dual auxiliaries "sein" (be) and "haben" (have), used, respectively, for verbs of motion or becoming (sein) or all other verbs (haben)? I've long assumed that the KJV-flavored English usage of "be" with "come" and the like is a throw-back to German usage; but that's just been speculation on my part.

    I particularly admired Tolkien's use of this flavoring in the emotive response he gives to his elf hero when confronted with his worst nightmare: "'Ai! ai!' wailed Legolas. 'A Balrog! A Balrog is come!'" This conveys, I think, much more feeling of impending implacable doom than would "A Balrog has come!", due to the Biblical feel imparted by "is come" (along with the reader's knowledge that elves are ancient in knowledge and are wont to speak in ways that call forth that ancientry).

  19. Ellen K. said,

    July 26, 2008 @ 2:52 pm


    Christ has died.
    Christ is risen.
    Christ will come again

    The idea is past, present, future. "Has died" points at the past. "Is risen" points at the present. It points at the present state of being risen, not the past act. "Has risen" would point to the past, not the present. (Speaking as a practicing Catholic here.)

    I'll leave it to others to analyze it grammatically, including whether it's relevant to the main topic here.

  20. Brett said,

    July 26, 2008 @ 3:33 pm

    @Steve Harris: This is drifting from anything linguistic, but I thought what made Legolas's dialog striking was the fact that the elf–who ordinarily seems to have ice water in his veins–is shrieking in terror.

  21. Danielle said,

    July 26, 2008 @ 3:55 pm

    There have always been perscriptivists… the educated and powerful always assume that their form of the language is the "right" way to speak. But there's also a phenomenon called "change from below" in which some changes in a language start as an "unpopular" or "incorrect" form that eventually becomes the standard. Relativity is a funny thing!

  22. Conrad H. Roth said,

    July 26, 2008 @ 6:17 pm

    Emile Benveniste has an interesting an essay on etre vs. avoir as auxiliaries: ""Etre" et "avoir" dans leurs fonctions linguistiques", in his 'Problemes de linguistique generale', pp. 187-207.

  23. Stephen Jones said,

    July 28, 2008 @ 3:30 am

    I think Mark, you are underestimating the petulance of Peevologists. They normally simply choose a particular form to excoriate, which is certain to be grammatical because otherwise nobody would be using it, and then make their absurd claims, normally based on class or regional grounds. If a particular form was not associated with a disliked social class or region then they simply wouldn't notice it.

  24. Kate said,

    July 28, 2008 @ 5:04 am

    I don't really have anything fascinating to add about this issue, but I enjoyed this post very much. Thanks!

RSS feed for comments on this post · TrackBack URI

Leave a Comment