Certain verbs, such as curro, have reduplicated perfect stems (such as cucurri). Other verbs, such as facio, fero had a reduplicated perfect stem in Old Latin (as seen on the Praeneste fibula) which were levelled by the time of Classical Latin.

Which verbs preserve the reduplicated perfect through Classical Latin? Are there any hypotheses in the literature as to why the paradigms of particular verbs were levelled and others were not?

  • 2
    I don't have a sourced answer: but I believe reduplication originated in PIE: Greek uses reduplication for almost all perfect tenses (luo -> leluka). It would be interesting to know why only a limited subset of Latin words does this. – brianpck Mar 24 '16 at 13:19
  • 1
    I also know of mordeo, momordi; posco, poposci. If I had to advance a hypothesis, it would be that all these actions which preserved a reduplicated perfect are quite energetic or vivid by nature. – jogloran Mar 24 '16 at 19:21
  • 1
    I wouldn't call this "paradigm leveling", since even after the loss of reduplication the perfect stem remains unpredictable and different from the present stem. "Inflection-class transfer" would be more accurate. – TKR Mar 25 '16 at 16:55
  • 1
    Further: disco, didisci; tango, tetigi; pello, pepuli; cado, cecidi; pango, pepigi. – jogloran Mar 25 '16 at 21:48
  • 1
    The Latin perfect is an amalgamation of two IE tenses: perfect and aorist. Perfect uses reduplication, aorist does not. – fdb Oct 1 '16 at 18:57
up vote 10 down vote accepted

Since you’re asking about reduplicated perfect (and not reduplicated present, as in bibo < *pi-ph3-e or sero < si-sh1-e, Weiss 2009: 405), I will try to address perfect formation only.

One of the problems is that synchronically we may not see all cases of reduplicated perfect in Classical Latin. However, by drawing on data from Old Latin and other Indo-European languages (especially from Greek, Sanskrit or Anatolian languages), we may discover more interesting cases.

For instance, Baldi mentions that

“Vowel weakening and the accompanying syncope have caused many Latin verbs with inherited PIE reduplication to lose the vowel of the reduplicated syllable” (p. 378).

Tronskii 1960 (2003) argues that if a reduplicated form had more than three syllables, the extra syllable was syncopated (para 624):

  • compound verbs: rettuli (*retetuli), reppuli (*repepuli), reccidi (*rececidi), repperi (*repeperi) etc.
  • cecidi but: occidi (*ob-cecidi), concidi (*con-cecidi);
  • tetigi but: contigi;
  • spopondi but: respondi;
  • tetuli but: abstuli (*abs-tetuli), attuli, sustuli etc.;
  • dedi or condidi but: abscondi;
  • converti (*conveverti);
  • incidi (*in-cecaidi) etc.

Tronski also mentions that tetuli was ousted by tuli, scicidi by scidi, tetendi by tendi etc. He also talks about some verbs that changed their perfect formation type because of syncope, e.g. tetini and continui etc.

Baldi also discusses Mester 1994, who examined the distribution of the three most common perfect formations. Baldi sums up by saying that perfect formation can be determined by prosodic, analogical, morphological and frequency factors.

Weiss 2009 writes that "verbs of second conjugation have u-perfect if the stem is light, but some other formation (s-perfect, reduplicated perfects) if the stem is heavy" (p. 413).

doceo - docui

BUT:

augeo - auxi

mordeo - momordi

He mentions a number of exceptions to this rule, though.

Now, if you really need a list of most common verbs that have reduplicated perfect, here's one, from Borovskii and Boldyrev 1975 - by no means exhaustive.

First conjugation:

do – dedi

  • circumdo – circumdedi

sto – steti

  • praesto – praestiti
  • obsto – obstiti

Second conjugation:

mordeo – momordi

pendeo – pependi

spondeo – spopondi

  • respondeo – respondi

tondeo – totondi

  • detondeo – detondi

Third conjugation:

cado – cecidi

  • occido – occidi

caedo – cecidi

  • occido – occidi

pendo – pependi

  • impendo – impendi

tendo – tetendi

  • attendo – attendi
  • ostendo – ostendi

credo – credidi

vendo – vendidi

perdo – perdidi

addo – addidi

  • condo – condidi
  • abscondo – abscondi

parco – peperci (parsi)

fallo – fefelli

  • refello – refelli

cano – cecini

curro – cucurri (cf. Old Latin cecurri)

  • accurro – accurri (accucurri)

posco – poposci

tundo – tutudi

  • contundo – contudi

tango – tetigi

  • attingo – attigi

pango – pepigi (pegi, panxi)

pungo – pupugi

  • interpungo – interpunxi

pello – pepuli

  • expello – expuli

disco – didici

sisto1 – (stiti)

sisto2 – steti

  • circumsisto – circumsteti
  • obsisto – obstiti
  • subsisto
  • desisto
  • resisto
  • existo

Third conjugation (io):

pario – peperi

Fourth conjugation:

comperio – comperi

reperio - repperi

As indicated in previous answers, reduplication comes from Proto-Indoeuropean, mainly for stative aspect and imperfective aspect verb forms. This reduplication is seen in many PIE-derived languages, as cited in the above article:

Ancient Greek: λύω lúō 'I free' vs. λέλυκα léluka "I have freed"
Gothic: hald "I hold" vs. haíhald (hĕhald) "I/he held"
Latin: currō "I run" vs. cucurrī "I ran" or "have run"

I cannot propose a convincing litmus test for which Latin verbs preserve reduplication and which ones do not. I did do some digging to find a full list of verbs with reduplicated perfect stems (since that was, after all, the primary question). Interesting to note: The vast majority are 3rd conjugation -o verbs. The only exceptions in the below list are do (1st), mordeo (2nd), and pario (3rd -io).

Verbs with Reduplicated Stems

(Please comment with additions/corrections so we can make a canonical list.)

cado, cecidit
caedo, cecidit
cano, cecinit
curro, cucurrit
do, dedit
disco, didicit
fallo, fefellit
mordeo, momordit (also: memordit)
pango, pepigit
parco, pepercit
pario, peperit
pello, pepulit
pendo, pependit
posco, poposcit
pungo, pupugit (also: pepugit)
tango, tetigit
tendo, tetendit
tondo, totondit (also: tundo, tutudit)

It is clear that the reduplicating verbs are deeply rooted in Proto-Indoeuropean, not only Latin and Greek exhibit them, but there are also traces of them in the Germanic languages (One class of strong verbs represented by German fallen, fällt, fiel, gefallen traces back to reduplicating verbs). Note that there was only a certain class that used reduplication, there were always other classes competing with them.

Latin was already in the stage of getting rid of reduplication: Only the simple verbs reduplicate, but prefixed forms (like incurro) don't. The force of analogy works towards the removal of residual reduplication, taking one verb after the other.

I am not aware of any rules governing this process and predicting what verbs are more conservative than others.

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.