Language Patterns
Recurring patterns across languages, space, and time.
Pathway · pathway

From a Result to a Past Tense

How does “I have it written” turn into an ordinary past tense?

Latin habeō litterās scrīptās means “I have the letter in a written state” — a description of how the letter is now. That sentence became French j’ai écrit la lettre, which reports the writing rather than the letter, and in speech it is now the plain past: the old simple past has gone out of use. The same road runs from Old Japanese -tari to modern -ta. English started down it and stopped: have written is still a perfect, and cannot be used with yesterday.

Domain
Tense and aspect
Structure
Auxiliary verb
Mechanism
GrammaticalizationReanalysisSemantic bleaching

Stages

  1. 1

    Resultative — a state something is in

    [HAVE/BE] + [OBJ] + [PTCP]

    The sentence describes how something stands now, and the participle agrees with it like an adjective. The earlier event is implied by the state, not asserted. Latin habeō litterās scrīptās, “I have the letter written”.

    On the timeline
    LatinOld EnglishJapanese
    Bybee et al. 1994Heine & Kuteva 2002
  2. 2

    Perfect — a past event that still counts

    [AUX] + [PTCP]

    The auxiliary and participle are reanalysed as one verb form reporting the event itself, with the result that it bears on the present. English I have written the letter, Spanish he escrito. In English and its neighbours the form resists a plain past-time adverb: not *I have written it yesterday.

    On the timeline
    FrenchEnglishGermanSpanishJapaneseSwedish
    Bybee et al. 1994Bybee & Dahl 1989Dahl & Velupillai 2013
  3. 3

    Past — the present relevance drops away

    [AUX] + [PTCP] = simple past

    The requirement of current relevance erodes, the form takes past-time adverbs, and it starts to crowd out whatever simple past the language had. Spoken French and southern German have gone this far.

    On the timeline
    FrenchGermanSpanishJapanese
    Bybee et al. 1994Drinka 2017
  4. 4

    A second layer, in some varieties

    [AUX] + [AUX.PTCP] + [PTCP]

    Where the old perfect has become a past, some varieties stack the auxiliary a second time: French passé surcomposé j’ai eu écrit, German Doppelperfekt ich habe geschrieben gehabt. These are anterior forms, largely confined to subordinate temporal clauses and to regional speech, and they have not replaced the stage-3 form. They stand outside the three-stage chain, and are shown here because they arise where it ends.

    On the timeline
    FrenchGerman
    Drinka 2017

Cross-linguistic evidence

Stages 1 to 3 are the chain set out by Bybee, Perkins and Pagliuca; stage 4 is a separate development that only some varieties show, listed here because it fills the gap stage 3 opens. Dates summarize the cited literature, band edges are approximate, and a language usually spends centuries with two readings available at once. Where a variety differs from the standard language, the note says so.

A state, then an event, then a past

The Latin sentence is about the letter and the condition it is in — litterae is plural in form but names a single letter, and the participle agrees with it. The French one is about the writing. Japanese arrived at the same place from 〜てあり, and modern -ta is simply the past.

Latin · lingua Latina

Habeō litterās scrīptās.

Natural
I have the letter in a written state.
French · français

J’ai écrit la lettre.

Natural
I wrote the letter.
Japanese · 日本語

手紙を書いた。

tegami o kaita.

Natural
I wrote the letter.

Geographic distribution

Each marker is one attested language, coloured by its evidence set; open it for the expression recorded there.

the same sentence, three stages the yesterday test building a new one Several sets

Marker positions are approximate cultural centres — they are not territorial claims. Tiles: OpenFreeMap · © OpenStreetMap contributors · Dark tiles: © CARTO.

Comparative historical timeline

How expressions overlap, compete, and replace one another across languages. Hover any band for the full date range and note.

-20002004006008001000120014001600180020002200LatinIndo-European › Italic (historical)habeō litterās scrīptās habeō litterās scrīptās (-100–400) Stage: Resultative — a state something is in A restricted construction, possible only where the subject holds the resulting state of a transitive object. The dominant past throughout the period is the synthetic perfect scrīpsī; this is the minor pattern that later takes over.FrenchIndo-European › Romancej’ai écrit j’ai écrit (1100–1600) Stage: Perfect — a past event that still counts Old and Middle French: a perfect beside the passé simple, which still carried plain narrative past.j’ai écrit j’ai écrit (1600–2000) Stage: Past — the present relevance drops away The passé composé takes over narrative past in speech; the passé simple survives in writing and formal registers only.j’ai eu écrit j’ai eu écrit (1600–2000) Stage: A second layer, in some varieties The passé surcomposé, regional and colloquial rather than standard, marks anteriority the passé composé no longer can.Old EnglishIndo-European › Germanic (historical)ic hæbbe þā bōc āwritene ic hæbbe þā bōc āwritene (800–1100) Stage: Resultative — a state something is in Old English habban with an agreeing participle; beside it bēon with participles of intransitive verbs. Uninflected participles appear in late Old English, and the reanalysis is under way before Middle English proper.EnglishIndo-European › GermanicI have written I have written (1100–2000) Stage: Perfect — a past event that still counts English settles here and stays: the perfect still refuses a plain past-time adverb, and no simple past has been displaced. American English goes the other way from stage 3 — the simple past has taken over ground the perfect holds in Britain, so “did you eat yet” answers what British English asks with “have you eaten yet”.GermanIndo-European › Germanicich habe geschrieben ich habe geschrieben (1200–1700) Stage: Perfect — a past event that still counts ich habe geschrieben gehabt ich habe geschrieben gehabt (1700–2000) Stage: A second layer, in some varieties The Doppelperfekt, in the same southern varieties that lost the Präteritum.ich habe geschrieben ich habe geschrieben (1600–2000) Stage: Past — the present relevance drops away The Präteritum is lost from speech across the Upper German south, and the Perfekt carries plain past. Northern German and written German keep the two apart.SpanishIndo-European › Romancehe escrito he escrito (1200–2000) Stage: Perfect — a past event that still counts Latin American Spanish keeps the perfect distinct and uses the preterite freely.he escrito (peninsular) he escrito (peninsular) (1600–2000) Stage: Past — the present relevance drops away The spread to events earlier the same day — he comido hace una hora — is under way by Early Modern Spanish and widens through the eighteenth century. Covering today is the step toward a plain past.JapaneseJaponic-tari -tari (800–1200) Stage: Resultative — a state something is in From -te ari, “having done, there is”: a state left by an event. In Old Japanese it is the newcomer, beside the established -tsu, -nu, -ki and -keri, and it grows dominant over the following centuries as those fall away.-tari / -ta -tari / -ta (1200–1600) Stage: Perfect — a past event that still counts Late Middle Japanese: -tari contracts to -ta and takes over the other past and perfective suffixes as they fall out of use.-ta -ta (1600–2000) Stage: Past — the present relevance drops away Modern -ta is the ordinary past and carries no requirement of present relevance. 〜ている has since grown into the language’s resultative, which is where -tari began.SwedishIndo-European › Germanicjag har skrivit jag har skrivit (1550–2000) Stage: Perfect — a past event that still counts The uninflected supine in -it separates from the agreeing past participle over the sixteenth century; before that, Old Swedish had hava with a participle that agreed, as Latin and Old English did. Like English, Swedish holds at the perfect and keeps a separate simple past.
How to read this timeline
  • Dominant use
  • Emerging use
  • Declining use
  • Approximate range
  • Overlap period (two forms coexist)
Stages
  • 1 Resultative — a state something is in
  • 2 Perfect — a past event that still counts
  • 3 Past — the present relevance drops away
  • 4 A second layer, in some varieties

References

  1. Bybee & Dahl 1989 01
    Bybee, Joan L.; Dahl, Östen (1989).
    The creation of tense and aspect systems in the languages of the world.
    Studies in Language. 13(1): 51–103. doi:10.1075/sl.13.1.03byb
  2. Bybee et al. 1994 02
    Bybee, Joan L.; Perkins, Revere; Pagliuca, William (1994).
    The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World.
    University of Chicago Press.
  3. Carey 1994 03
    Carey, Kathleen (1994).
    The grammaticalization of the perfect in Old English: An account based on pragmatics and metaphor.
    In Pagliuca, William (eds.), Perspectives on Grammaticalization (Current Issues in Linguistic Theory 109). John Benjamins.
  4. Dahl 1985 04
    Dahl, Östen (1985).
    Tense and Aspect Systems.
    Basil Blackwell.
  5. Dahl 2000 05
    Dahl, Östen (2000).
    Tense and Aspect in the Languages of Europe.
    Empirical Approaches to Language Typology (EUROTYP) 20-6. Mouton de Gruyter.
  6. Dahl & Velupillai 2013 06
    Dahl, Östen; Velupillai, Viveka (2013).
    The Perfect.
    In Dryer, Matthew S.; Haspelmath, Martin (eds.), The World Atlas of Language Structures Online, chapter 68. Max Planck Institute for Evolutionary Anthropology. https://wals.info/chapter/68
  7. Drinka 2017 07
    Drinka, Bridget (2017).
    Language Contact in Europe: The Periphrastic Perfect through History.
    Cambridge Approaches to Language Contact. Cambridge University Press. doi:10.1017/9781139027694
  8. Frellesvig 2010 08
    Frellesvig, Bjarke (2010).
    A History of the Japanese Language.
    Cambridge University Press.
  9. Harris 1982 09
    Harris, Martin B. (1982).
    The “past simple” and the “present perfect” in Romance.
    In Vincent, Nigel; Harris, Martin B. (eds.), Studies in the Romance Verb. Croom Helm. 42–70.
  10. Heine & Kuteva 2002 10
    Heine, Bernd; Kuteva, Tania (2002).
    World Lexicon of Grammaticalization.
    Cambridge University Press.
  11. Shirai 1998 11
    Shirai, Yasuhiro (1998).
    Where the progressive and the resultative meet: Imperfective aspect in Japanese, Chinese, Korean and English.
    Studies in Language. 22(3): 661–692. doi:10.1075/sl.22.3.06shi

Related patterns

Corrections

Any of this can be wrong. Dispute what the page says, add a language it is missing, or edit the file it was rendered from.

Dispute a claim Add a language Citation problem
Written in src/lib/data/pathways/resultative-to-perfect.ts Edit ↗ History ↗ How corrections work →