Excerpts from 'Linguistics across historical and geographical boundaries'
The function of prefixation in the assignment of
aspect to the Polish verb
Andrzej Pisarski
For almost a century now linguists have tried to deal with the
problem of verbal aspect, especially in the Slavonic languages. In
spite of the fact that enormous literature has been published on the
subject so far it is hardly possible to say that it has been satisfactorily
treated. There are valuable contributions to the description of the
Polish aspectual system […] but it seems that almost every one of
them leaves quite a number of doubts and problems that are not
convincingly accounted for. One such problem which still needs
further discussion and proper explanation is the problem of the
relation between a simple morphological form of the verb and its
prefixed formations. In a great number of treatments an opinion
prevails that prefixation of morphologically simple verbs, which are
imperfective, results in the formation of the perfective counterparts
of these verbs. Such observations can be found in almost every
Polish grammar handbook, be it a secondary school textbook or a
manual for a foreign learner of Polish. This opinion has been based
on the analysis of certain verbs such as:
pisać — napisać 'write'
robić — zrobić 'do, make'
czytać — przeczytać 'read' etc.
where the addition of a prefix to a morphologically simple stem indeed
results in the formation of the derived perfective construction.
Such an analysis would lead to a very convenient and very simple
generalization that prefixes added to imperfective verbs cause the
change into perfective verbs if it were not for the fact that the simple
verbs mentioned above can take a number of other prefixes, eg:
pisać — przepisać, dopisać, podpisać, wypisać, odpisać, etc.
robić — przerobić, dorobić, zarobić, porobić, etc.
czytać — wyczytać, odczytać, doczytać, etc.
It can be easily observed that each prefixed formation is perfective,
but at the same time each of these perfective counterparts of a
morphologically simple imperfective verb shows a subtle, sometimes
difficult to grasp, difference in meaning. This multitude of prefixed
forms derived from one and the same base causes that a satisfactory
interpretation and explanation of the process of prefixation is very
difficult to achieve. In fact, this multitude of derived forms corre-
sponding to one simple form of a verb seems to be the main source
of quite serious misinterpretations of the process of prefixation in
the assignment of aspect to verbs.
[…] Grammarians pointed out that
in very many cases the prefixation of a simple base with different
prefixes results in the apparent change of aspect from imperfective
to perfective. And yet, in quite recent morphological works one can
find statements to the effect that 'the basic function of a verbal
prefix is to express the perfective aspect of a verb. Each derivate in
which a prefix is a formant is always perfective, i.e. it informs that
the action is treated as complete unity, and besides quite often it
differs form the base verb by some semantic modification e.g. by
spatial direction, time location or quantitative characteristics of an
action.’ (Grzegorczykowa 1979:78).
Such a treatment of the process of the perfectivization of simple
Verb forms leads to still another problem, i.e. to the question of
the so-called ‘secondary imperfectives’. A number of publications on
the subject suggest that almost every prefixed, perfective formation
tends to form a further imperfective counterpart which can be treated as secondary in reference to its basic, simple imperfective form, e.g.:
pisać – przepisać – przepisywać
szyć – przyszyć – przyszywać
robić – dorobić – dorabiać etc.
However, not all the prefixed, perfective forms are capable of his
Further imperfectivization. Within each group of derived verbs there
is always, or almost always, one form which cannot be further derived
into a ‘secondary imperfective’. Consider the following examples:
robić – zrobić - *zrobiać
pisać – napisać - *napisywać
czytać – przeczytać - *przeczytywać
This also shows that within each group of verbs the prefixed con-
struction which is prevented from further derivation is not charac-
terized by any modification of meaning. In these cases the prefixed
formation differs from its base form by the aspectual function only.
The addition of a prefix denotes only the completion of an action,
or its natural termination. The usual explanation for this fact is in
terms of ‘empty prefixes’ which have purely aspectual function and
are free of any lexical meaning. The status of these empty prefixes
has never been satisfactorily explained. Only some of the linguists
admit that empty prefixation is the formal exponent of aspectual
oppositions (cf. Piernikarski 1969). Those against the existence of
lexically empty prefixes argue that one and the same prefix, added
to some base verbs, does cause the change of the lexical load of the
verb, and added to some other base verbs shows no lexical function,
as in:
robić – zrobić (empty)
jechać – zjechać (loaded)
pisać – napisać (empty)
sypać – nasypać (loaded)
Since it is difficult, or rather impossible, to establish the conditions
Under which a given prefix has no lexical load and under which it
Assumes some lexical function, it seems convenient to deny the
Existence of lexically empty prefixes. Yet this unexplained situation
Seems to bother quite a number of analysts.
An interesting interpretation was offered in Bogusławski (1963)
And later developed in Wróbel (1978). This interpretation lies at the
basis of the present analysis. Bogusławski’s basic observation was
that in the discussion and description of imperative and perfective
verbs and their mutual relations so far one cardinal error had been
made, i.e. the analysis had always been carried out in one direction
- from an imperfective base verb to a derived – prefixed –
perfective counterpart and, eventually to the still further derived
‘secondary’ imperfective in the case which had allowed such further
imperfectivization, i.e. in those cases where prefixes attached to
simple verbs had resulted in lexical modification or mutation. This
seemingly correct way of reasoning leads to the complicated situa-
tion in which one simple imperfective has sometimes a great number
of perfective counterparts. There are verbs in Polish which can take
all the sixteen verbal prefixes, not to mention the fact that they can
take combinations of prefixes. Thus a double function is attributed
to a verbal prefix – a) perfectivization of simple stem, b) modifica-
tion of lexical meaning. Instead, Bogusławski suggests that the
analyses should be carried out among pairs of verbs rather than
among single morphological form. Thus the establishment of the
function of prefixation should follow from the analysis of the
interrelations among such pairs as:
pisać/napisać : przepisać/przepisywać : wypisać/wypisywać, etc.
It now becomes clear that the prefix na- has purely aspectual
function only in the opposition pisać/napisać. The oppositions pisać/
przepisać, pisać/wypisać, etc. show only semantic modification of
the base verb. The fact that prefixed verbs happen to be perfective
is only incidental. This was pointed out elsewhere and much earlier
(Agrell 1918: 4 and Koschmieder 1934: 18), but these observations
Seemed to have been largely neglected by other aspect analysts.
The fact that some prefixes, such as na- and z- mentioned above,
Sometimes have lexical function and sometimes are only empty
morphemes has traditionally been explained by an argument that
verbal prefixes, as opposed to suffixes, have a double, if not mani-
fold function. Instead, Bogusławski and Wróble suggest that we
have to do not with a double function of the prefix but with two
independent sets of prefixes, one lexical and one purely aspectual.
Some of the prefixes within each set happen to be homophonous
and it is the homophony that causes the whole apparent complexity
of the problem of aspect. Whether the prefix attached to the base
verb has aspectual or lexical function should be established on the
basis of a semantic analysis of the verb, and not morphologically.
On the basis of the observations of both Boguławski and Wróbel
as well as the earlier arguments of Agrell and Koschmieder it can
be claimed that the two above-mentioned sets of prefixes operate
on two different levels in the lexicon. The morphemes which have
traditionally been referred to as lexically loaded prefixes operate on
the level of a very specifically understood process of word formation
- they serve the purpose of modifying the basic verbal meanings.
In other words, lexical prefixes are those elements which operate
on basic verbal lexemes in order to form new, modified lexemes.
E. g., do- when added to pisać writing’. At this level verbal lexemes,
whether basic or modified, whether simple or derived, are indifferent
as far as their aspectual function is concerned. Aspect, either perfec-
tive or imperfective, is assigned to lexemes only after the modifica-
tion of meaning has been completed (cf. Wróble 1978: 114ff.)
|
![Start a new thread in [General Language]](http://www.polishforums.com/img/poland-thread.jpg "Start a new thread in [General Language]")