* * * * * * * * * * * * * * * * * * * * * * * *
LINGUIST List logo Eastern Michigan University Wayne State University *
* People & Organizations * Jobs * Calls & Conferences * Publications * Language Resources * Text & Computer Tools * Teaching & Learning * Mailing Lists * Search *
* *
LINGUIST List 18.68

Wed Jan 10 2007

Sum: Usurpative Etymology of Suppletive Forms

Editor for this issue: Kevin Burrows <kevinlinguistlist.org>

To post to LINGUIST, use our convenient web form at http://linguistlist.org/LL/posttolinguist.html.
        1.    Konrad Szczesniak, Usurpative Etymology of Suppletive Forms

Message 1: Usurpative Etymology of Suppletive Forms
Date: 10-Jan-2007
From: Konrad Szczesniak <konrad.szczesniakgmail.com>
Subject: Usurpative Etymology of Suppletive Forms

Query for this summary posted in LINGUIST Issue: 17.2948
Regarding Query: http://linguistlist.org/issues/17/17-2948.html#1 

Dear Colleagues,

This is a long overdue summary of the responses to our query on
suppletion from almost four months ago. The query originally focused on
the distinction between 'usurpative' and 'fusional' suppletion. Usurpative
suppletion was interpreted as including (in an otherwise regular paradigm)
forms usurped from other lexemes. On the other hand, 'fusional' suppletion
was assumed to involve combining forms from more than one lexeme,
resulting in the complete absorption of the donating lexemes.

The literature and the examples sent in by many LL subscribers showed that
instead of being a clearcut dichotomy, the above distinction is more of a
continuum. It has also become clear that despite over a century of studies,
suppletion is still a challenging question. The origins and the exact
mechanisms of the formation of suppletive forms are unclear. It has also
been controversial how suppletion should be defined, and what phenomena
should be considered part of it.

Although the definition of suppletion is often restricted to inflectional
suppletion, with verbal suppletion being the most studied case, derivational
suppletion is also argued to be possible. However, it should be noted here
that derivational suppletion is a controversial category. When cases of
derivational suppletion are found, they come from derivational types that
are typically adduced to illustrate the fuzzy borderline between derivation
and inflection, like morphological means for marking gender. Common
examples are gender pairs like 'byk / korova' (Russian 'ox / cow') or 'baran /
owca' (Polish 'ram / sheep'). Unfortunately, gender marking is among a
number of problematic processes (including adjectival/participial
morphology, etc.) straddling the line between inflection and derivation. In
other words, even when derivational suppletion is shown to exist,
paradoxically the examples given seem to prove the inflectional nature of

We are not aware of cases of suppletion for 'fully' derivational processes like
nominalization or antonymy; what's more, we would like to venture that
suppletion is impossible in true derivation. It is ruled out by the defining
feature of derivation, ie. semantic opacity which makes it impossible to
propose uniform derivational paradigms. For example, most verbs which
convey the idea of repeated action are formed by adding the prefix re-, as
in 'restate', 'reheat', 'retell', 'rewrite', 'redo' or 'replay'; here it would possible
to think of suppletive forms like 'echo', 'duplicate' or 'copy'. The problem
with these examples is that it is not altogether clear that re- forms belong
under a uniform semantic paradigm which would justify searching it for
'suppletive inclusions'. Because derivation always involves some semantic
opacity, the idea of repetition is not the only semantic feature which sets
derived verbs apart from their original forms. For example, 'rewrite' does
not mean write (the same letter) again. Similar (or perhaps even more)
semantic inconsistency is notorious in antonymy. Because of such
irregularities, no 'derivational paradigms' are available for accommodating
suppletive forms.

Other problems arise regarding definitions of suppletion based on
etymology. According to Rudes (1980), not all types of extreme irregularity
are suppletive enough. For example, the Polish pair 'ciac / tne' (to cut,
infinitive / 1SG) is termed 'pseudo-suppletive', because the two forms are
etymologically related. This approach is motivated by the assumption that
suppletion is exemplified by cases where a paradigm contains a ''foreign
body'' coming from other lexemes. This position is dismissed by Mel'čuk
(1994) and Veselinova (2006, the most recent large-scale study of
suppletion and a review of previous work), offering a practical reason for
dismissing etymological considerations in defining suppletion: ''a definition
based on etymology makes a typologically oriented study practically
impossible as historical information of this kind is not available for many of
the world languages (Martin Haspelmath, p.c.).''

Thus, Veselinova identifies three sources of suppletion: 1.)
grammaticalization; 2.) lexicalization, and 3.) the loss of regular
morphological processes involving forms of a single lexeme, resulting in
the desintegration of its paradigm. Veselinova shows that suppletive forms
tend to occur in lexemes which are becoming grammaticalized, or in
lexemes subject to semantic change dynamics.

Interestingly, the three sources converge on what they affect; suppletion
tends to affect high-frequency lexemes like 'be', 'go', 'sit', 'take', or 'say'. In
languages with verbal suppletion, it is these verbs that universally exhibit
suppletive forms. However, as Veselinova notes, the question of frequency is
quite mysterious. It is not a sufficient condition for suppletion, as there are
frequent verbs without suppletive forms (e.g. 'think'), whereas some less
frequent verbs (like 'put' in Slavic languages) are prolific hosts of suppletion.

A review of examples of suppletion raises cause-or-effect doubts regarding
the role of donating lexemes – does suppletion really result in the
absorption of donating lexemes or is suppletion facilitated by their loss of
autonomous lexeme status? The scenario of the development is addressed
in Rudes (1980). He observes that ''[t]he mechanisms by which suppletive
verbs come into existence have never been well understood , principally
because no case of incipient suppletion has ever been noted and studied in
detail. In most languages, those suppletive verbs which exist are of ancient
date and there are no records of their creation.'' But Rudes demonstrates
what might be a case of suppletion formation underway: the fusion of two
Romanian verbs 'a vrea' and 'a voi', the former being irregular and the latter
a regular donor of forms which enter the paradigm of 'a vrea'. Rudes shows
that while some forms of the irregular 'a vrea' are still usable, they are
gradually losing out in frequency and are being supplanted by the regular
forms from the 'a voi' paradigm. Although still separate, the two paradigms
are hypothesized to fuse at some point. According to Rudes, irregularity
exposed to acquisitional pressures (children do not acquire low-frequency
forms which are eventually replaced by regular forms from other
paradigms) may facilitate the formation of suppletive paradigms.

Other major questions that remain to be answered include why some
languages exhibit suppletion while others seem ''immune'' to it. Sometimes,
the answer is obvious: for example, in languages like Chinese, there is little
to supplete; there are no inflectional paradigms to supply with ''foreign
matter''. But on the other hand, one needs to explain the conspicuous
absence of suppletion in many cases of inflected languages (for example,
no suppletion for number in Polish).

Finally, an interesting leftover question is whether 'kill / die' should be
considered a suppletive pair. In many languages, the two are unrelated
verbs which were famously demonstrated by Fodor (1975) to be
syntactically separate beings without a derivational link, an observation
confirmed by the intuitions of native speakers of lanaguages where the two
are separate. However, in some languages (like Basque), the two are
expressed by two related forms participating in the causative alternation,
indicating that some kind of causative-inchoative relation must hold
between them and therefore categorizing them as suppletive is perhaps not
entirely unjustified.

We would like to thank Ante Aikio, Susan Fischer, Matt Juge, John Koontz,
Mary Marino, Marc Picard, Blair Rudes, Hayim Y. Sheynin, Herb Stahlke, Péter
Szigetvári, Ljuba Veselinova, and Ghil`ad Zuckermann for examples and
their comments.

Konrad Szczesniak
Silesian University
Sosnowiec, Poland

Marcus Callies
Philipps-Universität Marburg
Marburg, Germany


Fodor, J. A. 1970. Three reasons for not deriving 'kill' from 'cause to die'.
Linguistic Inquiry 1: 429-438.
Mel'čuk, I. 1994. Suppletion: Toward a logical analysis of the concept.
Studies in Language 18(2): 339–410.
Rudes, B. 1980. On the nature of verbal suppletion. Linguistics 18: 655–76.
Veselinova, L. N. 2006. Suppletion in Verb Paradigms. Amsterdam: John
Benjamins Publishing Company

Linguistic Field(s): Typology

Respond to list|Read more issues|LINGUIST home page|Top of issue

Please report any bad links or misclassified data

LINGUIST Homepage | Read LINGUIST | Contact us

NSF Logo

While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.