Philosophische Fakultät der Universität Zürich

Lizentiatsprüfungen SS 2000

Dreitägige Hausarbeit

Thema:

Aspect in Old English

Fach: Englische Sprachwissenschaft Marc Zumstein

Prüfender: Prof. Dr. Andreas Fischer Waffenplatzstrasse 91
Datum: 22.02.2000 – 25.02.2000 8002 Zürich
Aspect in Old English Marc Zumstein

CONTENTS

0. Introduction ........................................................................................................................................ 1

1. Initial remarks on ge- .......................................................................................................................... 2

2. A first approach – the current doctrines ............................................................................................ 3

2.1. The situation in some handbooks of Old English ........................................................................ 3
2.2. LINDEMANN’s “re-examined doctrines” ......................................................................................... 6
2.3. The levels of argumentation ........................................................................................................ 8

3. Some attempts to classify predicates ................................................................................................. 9

3.1. VENDLER’s classification ................................................................................................................. 9
3.2. Quirk’s classification .................................................................................................................. 10
3.3. BULYGINA’s classification.............................................................................................................. 10
3.4. COMRIE’s considerations ............................................................................................................. 10

4. Understanding and misunderstanding aspect ................................................................................. 11

5. Two competing ideas about aktionsarten ........................................................................................ 13

6. Functions and effects of prefixation ................................................................................................. 16

6.1. Germanic languages................................................................................................................... 16
6.2. The Slavic prefix по-/po- ............................................................................................................ 17

7. A new theory about ge- .................................................................................................................... 18

7.1. Some predecessors .................................................................................................................... 18
7.2. LINDEMANN’S new theory ............................................................................................................. 19
7.3. Why did ge- disappear? ............................................................................................................. 20

8. The expanded form in Old English .................................................................................................... 21

9. Summary ........................................................................................................................................... 22

10. References....................................................................................................................................... 24

Page 2 of 27
Aspect in Old English Marc Zumstein

0. INTRODUCTION

An older word for ‘verb’ in ModHG, Zeitwort, literally ‘time word’, draws our attention to a category
that is obviously perceived as the most prominent for verbs, time. In those languages that we are
familiar with, the semantic concept of time reference is expressed by the grammatical category of
tense. Comrie describes the basic function of tense as follows: “Tense relates the time of the situation
referred to to some other time, usually to the moment of speaking” (1976: 2). Other languages, how-
ever, rely to a larger or lesser extent on another concept to express time reference, a concept com-
monly referred to as aspect. In fact, there are even languages completely lacking tenses, thus working
exclusively with aspect. Even though it is time that both tense and aspect focus on, there are sub-
stantial differences in the way they do so. To outline this very briefly, aspect, unlike tense, does not
relate the time of a given situation to any other time point, it is rather concerned with different ways
of viewing the internal temporal constituency of the one situation. “One could state the difference
as one between situation-internal time (aspect) and situation-external time (tense)” (ibid.: 5). Aspect
as a grammatical category is much less overt than tense. First of all, in different languages there are
different notions of aspect, but quite commonly they form binary aspectual oppositions meaning that
the aspectual value of a verb is determined only through its counterpart with the opposite value1.
Another important feature of aspect is its dependence on the surrounding parts in the sentence and
even more on the semantics of the verb or more precisely verb phrase. Aspect can be marked mor-
phologically, e. g. the English -ing – form together with a form of the verb to be that as markers of
the progressive aspect, or through periphrasis, as we see it in the French construction être en train
de faire quelque chose, with a notion very similar to the English progressive aspect.
Following COMRIE, who shall be our theoretical guide in this paper (although there are some points of
objection), the following issues in connection with aspect will be of interest: the perfective – imper-
fective opposition, the progressive – nonprogressive opposition, the notion of habituality, the English
perfect tense, the relation between prefixation and aspect and, last but not least, the relation be-
tween inherent meaning and aspect. The topic of this paper, Old English, allows two “aspectual ap-
proaches”. On the one hand, there is the diachronic approach comparing Old English to Modern Eng-
lish with its grammaticalized progressive – nonprogressive opposition. This issue, however, will only
be treated briefly at the end of the paper, since there is aн almost general agreement that the use of
Old English “progressives” like in mid þi þe he þas þingc wæs sprecende to him silfum2 was only a
stylistic variation, even though sometimes it fits into the ModE pattern perfectly.
On the other hand, there is an approach we might call typological, namely the theory that there was
in Old English an aspectual distinction evoked by the desemantized verbal prefix ge-, postulated in
more or less close analogy to the system in the Slavic languages. My guiding question will be: How
can we account for this theory to have emerged and how can it be justified, if at all? This includes the
“subordinate” questions whether there was a verbal aspect in Old English in first place or how it was
expressed formally. This starting point sets the determination of meaning and function of ge- as the
central issue when discussing Old English aspect.

1
An illustration from Russian: from the verb стрелять (streljat’) ‘shoot’ two compounds can be derived: застрелять
(zastreljat’) and расстрелять (rasstreljat’), both meaning ‘shoot at (somebody)’. They look absolutely identical, but still
the first belongs to the imperfective aspect, because it has the perfective counterpart затрелить (zastrelit’), whereas the
second pairs with an imperfective, расстреливать (rasstrelivat’), and is thus perfective.
2
The example is cited from the Old English translation of the Apollonius tale.
Page 1 of 27
Aspect in Old English Marc Zumstein

1. INITIAL REMARKS ON GE-

The main focus of this first part of my paper is to deal with a language phenomenon of Old English
that should be familiar to anybody who has ever been exposed to this language, first of all countless
students of the English language at universities all over the world. In Language: its structure and use,
FINEGAN describes a basic principle in the functioning of language as follows: “In the circumstances of
human communication, our ideal systems represent competing poles between a need for explicitness
and a drive toward efficiency” (1994: 3). In other words, any language is rather unlikely to tolerate
elements within its system that do not contribute to the transmission of information in one way or
another, or, if we put this in a somewhat simpler way, that do not carry meaning. “A language can be
viewed as a coin whose two sides are expression and content” (ibid.: 4). But at first glance, we find
this principle violated in Old English. How do we account for the glossing in A Guide to Old English by
MITCHELL/ROBINSON, e. g. both sittan and ġesittan with ‘sit’ or þenċean and ġeþenċean ‘think’? Is this
preverb ge- just a luxury of language, an expression without content, semantically negligible? MITCH-
ELL/ROBINSON support this view, when they write in the introduction to their glossary that “the prefix
ġe- is ignored in alphabetizing words”, a practice which, by the way, is quite representative also for
other dictionaries of Old English. But nonetheless, verbs with this prefix still appear in the glossary,
thereby denying the possibility that they were mere parts of the inflectional system. A review over
the explanations of ge- offered in some textbooks of Old English will be presented in the following
chapter.
The question about the meaning or the function of Old English ge- and its equivalents in the other
Old Germanic dialects, like ga- in Gothic or gi- in Old High German, was treated in many monographs
and articles that were published in the second half of the 19th and the first half of the 20th century3.
Although the results of these works were controversial and often heavily disputed, the discussion of
ge- seems to have lost interest in today’s studies of English, the last contribution known to me being
LINDEMANN’s revision of current doctrines about the preverb in 1965 and his own theory about its
meaning in 1970.
Undoubtedly a problem like this must be examined on the basis of empirical data, and not only of Old
English material, but also taking into account the parallel occurrences in other Old Germanic dialects,
a procedure which of course would exceed the limits of this paper by far. Additionally, such empirical
data would still need to be relativized, as with dead languages we do not have any native speakers to
verify conclusions out of these data.
Still this paper sees its task not only in gathering and summarizing what others have written. Apart
from a close analysis of the two main theory-groups4 about ge-, the main issue will be to produce a
synthesis of these theories within a broader grammatical and semantical framework.
Another effort shall be made to point out doubtful uses of linguistic terminology in the works of some
scholars as well as misunderstandings and misinterpretations they led to in the works of others. In
order to understand the proposed framework leading to our final synthesis, a substantial part of the
statements to follow will be analyses of grammatical and semantical models that have been used in
the past to explain the use of the preverb ge-, the most prominent of them being the notion of aspect

3
The bibliography in LINDEMANN 1970 gives a detailed overview over these publications. Contrary to E. G. STANLEY in his
discussion of LINDEMANN 1970, however, I do not see this bibliography as “the most valuable contribution in this accurately
printed book” (STANLEY 1973: 494).
4
Following the principal line of my argumentation, I divide the authors on this subject into those that operate with the
term aspect and those that decline to do so. Similarly, LINDEMANN 1970 uses the opposition between the doctrine insisting
that the morpheme ge- was a marker of aspect and an explanation of its original lexical meaning as a basis for discussion.
(vii)
Page 2 of 27
Aspect in Old English Marc Zumstein

and a concept called aktionsart5. As the controversy between the two main doctrines is closely linked
with the situation in the Slavic languages, namely one group of scholars drawing parallels between
the Germanic and the Slavic languages in connection with assumed aspectual distinctions evoked by
ge-, and the other one accusing the first one of wrongly doing so, some further clarifications from
this part will also need to follow.

2. A FIRST APPROACH – THE CURRENT DOCTRINES

After a short overview over the way the meaning or function of the prefix ge- (or preverb, as some
prefer to call it) is described in some textbooks on Old English, the five “wrong doctrines” LINDEMANN
(1965) lists will be presented, which can be regarded as a basis for our further discussions. Interest-
ingly enough, we can find all of these doctrines already in the textbooks. Finally, I will also briefly talk
about the levels of argumentation that I will use for my revision and reinterpretation of what has
been written so far.

2.1. The situation in some handbooks of Old English

Books with the aim to teach a language have to communicate the essential knowledge needed to
understand the system of that language. I therefore consider a glance at the way Old English hand-
books handle this problem a reasonable first approach. The crucial question at the beginning of such
an examination will be where exactly a given handbook deals with ge-, thereby already revealing
much of the author’s understanding of it. Basically, there are three possibilities, namely inflectional
morphology, syntax and word formation. When ge- is treated as a part of inflectional morphology
and/or syntax, we can conclude that we are confronted with an aspectual reading, whereas deriva-
tional morphology and/or word formation state the addition of a non-aspectual semantic notion.
Very often, however, these two ways of description are also found to be mixed up.
Nearly all of the six textbooks consulted, namely PILCH 1970, PINSKER 1976, DÜRMÜLLER/UTZ 1977, MITCH-
ELL/ROBINSON 1992, BARBER 1993 and WEIMANN 1995, discuss ge- as a part of word formation, the sole
exception being PINSKER 1976, where the only mentioning of ge- can be found in his word index, a
plain statement that this prefix originally meant ‘together’. Four of the other explanations use the
term “perfective”, whereas the fifth explicitly rejects it. It might now be worth taking a closer look at
these, and first of all at their use of “perfective”, which also differs among the various authors. Thus
DÜRMÜLLER/UTZ (1977: 50) discuss ge- within morphology and write that in earlier periods of Ger-
manic, this prefix (or ga-, as it appeared in Gothic) indicated a completed6 situation. Further they
write that in Old English, ge- was a part of the preterite participle, where it emphasized the expression
of perfective aspect. From the synchronic point of view, this limitation to the preterite participle can
hardly be accepted for Old English7, even though DÜRMÜLLER/UTZ relativize this by adding that ge- is

5
COMRIE writes about the term aktionsart, that “although there have been numerous attempts to coin an English equiv-
alent, none of these has become generally accepted” (1976: 6f). The Metzler Lexikon Sprache, for example, proposes
lexical aspect. For the sake of convenience, I will continue using the term aktionsart.
6
It is important to keep in mind that there is a significant difference between completed and complete situations; I will
return to this topic in chapter 4.
7
This is rather the state of affairs of ModHG, where “ge- is simply a regular marker for the Past Participle” (COMRIE 1976:
91).
Page 3 of 27
Aspect in Old English Marc Zumstein

also found in various other verb forms, “because it had become a very productive element of word
formation”8.
It will be argued later in chapter 7.3. that I believe this development to have happened in the opposite
direction; now we first have to get even with the use of the term perfective aspect. In English linguis-
tics, there has been a very specific use of this term, quite different from the meaning ascribed to it in
general and first of all in aspect linguistics. So one of the standard grammars of English, QUIRK et. al.
1985, first defines “aspect” as a category that distinguishes the way a described verb action is re-
garded or experienced (ibid.: 188). Then, somewhat inconsequentially, from my point of view, the
term perfective aspect is introduced, which actually has nothing to do with “aspect” in the strict sense
of the word, as it simply indicates anterior time (ibid.: 190). Undoubtedly the English perfect tense(s)
and the perfective aspect do share some common features, which will be discussed in chapter 4, but
nonetheless it still cannot be disputed that we are talking about tenses, and not about aspect. COMRIE
is well aware of this; he writes: “Aspect (…) [is] concerned with different ways of representing the
internal temporal constitution of a situation. The perfect is rather different from these aspects, since
it tells us nothing about the situation in itself, but rather relates some state to a preceding situation.”
(1976: 52) I go into this with some detail, because the confusion sometimes is quite striking. Even
MITCHELL in his monumental Old English Syntax does not get along with it. In an attempt to criticize
an earlier work on this subject by QUIRK and WRENN, where on the one hand perfective aspect refers
to the function of the English perfect tense9 and on the other hand the term perfective is used to
describe what I will explain as “telicity” in chapter 3, he fails to reveal the underlying concepts by
analyzing the confused and confusing terminology (1985: 365f.). Also, one cannot be sure from his
discussion that he is familiar with the differences between “real” perfective and perfectic10.
It is evident from their further discussions that DÜRMÜLLER/UTZ connect preterite participles plus ge-
with a perfectic meaning, thus denoting anterior time reference. Since participles are non-finite verb
forms, they already contain what COMRIE calls “relative tense” meaning that they “relate the time of
the situation described to the time of some other situation” (1976: 2). When DÜRMÜLLER/UTZ (1977:
92) return to the ge- question in connection with lexicon, they mention many effects of compounding
with ge-. They observe that ge- can form transitive verbs out of intransitive simplexes. So it is said
about the Danes that they geridon and gesæton Westsæxna land, where we find the opposition ridan
‘to ride’ and sittan ‘to sit’ versus geridan ‘to raid’ and gesittan ‘to occupy’. Also a certain change in
meaning cannot be neglected. But this is not a principle that can be generalized, so DÜRMÜLLER/UTZ
presume that the additional semantic elements evoked through compounding are an intensifying of
the action, an emphasis of engagement (when used in an imperative) or a result, that “could be the
intention of ge- with verbs” (ibid.). To exemplify this notion of result11, they cite the well-known ex-
ample of winnan ‘to fight’ and gewinnan ‘to win’. But as this explanation can be valid also only for a
very limited number of verbs, DÜRMÜLLER/UTZ further propose synonymy of simplex and ge- com-
pounds, by means of which they also explain the “remarkable fact that the derivational prefix ge- has
got lost almost completely in ModE” (ibid.). However, such synonymy may cause certain conflicts
with sign theory, as I have pointed out in the introduction, and so DÜRMÜLLER/UTZ conclude that “the

8
The emphasis is mine.
9
QUIRK and WRENN argue that this function could also be carried out by the Old English “simple preterite”, which is of
course not wrong, since the compound perfect forms were at best only emerging at that time.
10
Following for example BLOOMFIELD 1929, I will use the term “perfectic” when referring to the notions conveyed by the
English perfect.
11
As shall be shown, this “resultativity” is a crucial point, because it is here where also the concepts perfective and per-
fectic meet to some extent.
Page 4 of 27
Aspect in Old English Marc Zumstein

function of ge-, which virtually overgrows the Old English lexicon, is one of the most disputed ques-
tions in Old English linguistics”12 (ibid.).
Similar statements, although not as detailed, are given by WEIMANN (1995: 104f.), namely that the
prefix is often desemantized (implying that simplex and compound may be regarded as synonyms) or
that it has a perfective function (also with the ambiguous reference to the preterite participle). The
same situation we find in MITCHELL/ROBINSON: “In verbs, it [ge-] sometimes has a perfective sense, e.
g. ge-ascian ‘find out’ and ge-winnan ‘get by fighting, win’; hence its frequent use in past participles”
(1992: 58). From all the handbooks consulted, only BARBER seems to have an understanding of per-
fective distinct from the ModE perfect, though not fully compatible with aspect theory: “Another
common verbal prefix is ge-, which often has a perfective force, signifying the achievement or the
completion of the action” (1993: 121). He cites “a well-known example of this perfective use of ge-
in King Alfred’s account of a voyage by the Norwegian Ohthere: þa siglde he þonan suþryhte be lande
swa swa he mehte on fif dagum gesiglan”, where “the interesting thing (…) is the difference between
siglde ‘sailed’ and gesiglan ‘to get somewhere by sailing’.” This idea does have some truth in it, but
still it requires some modification as far as the use of perfective and the concept used in this particular
context are concerned.
The last work consulted, PILCH 1970, glosses the prefix ge- as ‘attain something’ plus rendering an
intransitive verb transitive. So we have, apart from the inevitable winnan – gewinnan pairs like gan
‘to go (itr.)’ – gegan ‘to conquer (tr.)’ or fon ‘seize (itr.)’ and gefon ‘catch (tr.)’13 (126). Even though
his notion of ‘attaining something’ corresponds very well with what the other authors refer to as
perfective, still PILCH, similarly to PILCH 1953, cannot refrain from polemizing against those scholars
who “following W. STREITBERG, saw the verbal derivation by means of prefixes as an equivalent to the
grammatical category of aspect or aktionsart in the Slavic languages, namely simplex verbs corre-
sponding to imperfective and compound verbs to perfective forms. This doctrine has meanwhile
turned out to be a misunderstanding” (ibid.). With all due respect, I still believe the misunderstanding
to be on PILCH’s side. It is not quite clear whether he just followed a tradition of polemizing (there is
some evidence for such a tradition), i. e. polemizing against this particular terminology, or whether
he actually challenges the underlying concepts. This second possibility becomes less likely in the light
of the following: First of all, as will be discussed later in chapter 5, STREITBERG did not postulate such
an equivalence for Gothic, on the contrary he explicitly denies it14. It is quite a different issue that he
uses certain parallels to exemplify his ideas. Second, I do not believe that PILCH really made an effort
to understand either STREITBERG’s theories or the precise situation in the Slavic languages, because in
PILCH 1953, he “compares apples with pears” to defeat STREITBERG (131f.)15.

12
With reference to noun compounds with ge-, DÜRMÜLLER/UTZ and MITCHELL/ROBINSON (1992: 58) follow PINSKER 1970 in
regarding ge- as the equivalent, genetically as well as sematically, of Latin con- ‘together with’, with examples like gesprec
‘discussion, con-ference’ (cf. ModHG Gespräch), or gebroþru (masc. plu.) ‘brothers’, Latin con-fratres. WEIMANN (1995:
104f.) calls this the original collective function of ge-.
13
Glossings like these make sense and surely reflect the given situation to a large extent. Another glossing of the same
author, however, shows that we cannot accept such translations as pure truth, but that we always have to calculate a
certain margin of incertitude, which one might be tempted to use as space for one’s own interpretations: don ‘to do’,
gedon ‘to make’ (PILCH 1955: 52).
14
STREITBERG criticizes a Czech linguist who wrote about the value of certain prefixes in Gothic: “The basic assumption of
BEER is wrong, because despite all warnings he classifies the Gothic structures according to the elaborated system of
Slavic, thereby not realizing the specific characteristics of the Germanic development.”
15
Similarly, there is a difference between aspect and aktionsart, and STREITBERG only meant the latter, despite of LINDE-
MANN’s hypothesis that STREITBERG used it as a (mis)translation Slavic vid ‘aspect’ (1969: 262f.).
Page 5 of 27
Aspect in Old English Marc Zumstein

2.2. LINDEMANN’s “re-examined doctrines”

In his revision of LINDEMANN 1970, STANLEY – not without a little irony – writes about the first part (“A
Re-Examination of Some Current Doctrines”) that “it would have to be a very good article indeed to
stand so much rereading”16 (1973: 493). As a matter of fact, it “leads us no further than understand-
ing what [ge-] does not mean” (LINDEMANN 1969: 259). The author sees his main task in listing existing
explanations of ge- and proving that and why they are not valid general theories to account for the
whole system of Old English. He does not, however, take into consideration the possibility that each
of these doctrines represents partly the “truth”17, meaning that together they might be joined to
form a general theory; nor does he do so in the second part on “The Meaning of the Morpheme”. His
starting point is clear: “The prominence of the preverb is such that it cries out for resolute endeavour
on our part to understand it” (ibid.). No less clear is the fact that there have already been numerous
attempts to determine the meaning and the function of ge-, but that they have been confused and
therefore unsuccessful. LINDEMANN explains this “obscurity”, as he calls it, in two approaches. On the
one hand, it is due to the very high degree of frequency of the preverb ge-, which results in a lower
emphasis in spoken language and may therefore lose a specific denotation and acquire a very general
one or several less specific and very general ones. On the other hand, he believes the more important
and immediate reason for the obscurity to lie “in the nature of the very doctrines that have been
proposed to explain the morpheme” (258). He regards theses as “philosophical hypotheses” rather
than “strictly linguistic observations confined to a homogenous language system”. He further argues
that evidence brought up to substantiate any of these doctrines will result from data that have been
especially selected, so that the doctrine in question will “reveal a very seductive cogency” (ibid.). A
glance at those few examples mentioned above easily illustrates this, e. g. the doctrine of evoking
transitivity and the difference between gan and gegan or ridan and geridan. To summarize this in the
language of logic, it can be stated that LINDEMANN begins his discussion with the antithesis, in the
second part he will add the thesis, but he fails to combine them to a synthesis. On the other hand, he
definitely gives a good insight into what has been thought and written so far, thus serving as an initial
point for further studies on this matter.
All of the following doctrines are already familiar to us, as all of them have been mentioned at least
once when discussing the situation in the Old English textbooks. They shall now be briefly repeated,
always together with LINDEMANN’s main arguments against them.

Doctrine 1: ge- is without meaning

The theory that the preverb ge- is desemantized, that it does not even have a grammatical function
is obviously very old. LINDEMANN cites the Vocabularium Anglo-Saxonicum of 1701 by Thomas BENSON
of Oxford, who wrote: “ge- apud Saxones semper fere superfluum”. I agree with LINDEMANN that the
fact that many of the already quoted simplex – compound oppositions have rather different mean-
ings is self-explanatory. Though a point worth being criticized is his treatment of synchronic and dia-
chronic appearances in language. Any language is the result of a process, of an evolution and itself
constantly involved in this process, and it must be clear that the system of a language is created by
human thinking, whereas the language described by this system is usually a natural phenomenon.
Elements of language that do not fit the patterns of such a system are referred to as “irregular”, but
such irregularities usually are remnants of an earlier period of that language, a period where they

16
After publication in JEGP 64/1965 (pp. 65 – 84), it was reprinted in the version cited in this paper and then again as Part
I of Lindemann 1970.
17
Under the condition, of course, that there is such thing as “truth” in this context.
Page 6 of 27
Aspect in Old English Marc Zumstein

were regular. Therefore a particular language element can not necessarily be explained synchroni-
cally, even though if ge- was actually meaningless, its high frequency in language would not allow to
regard it a pure remnant of Proto-Germanic. But even the time difference between Old English and
Gothic, which I estimate as roughly 500 years, leaves room for many changes and differing develop-
ments, which is why I question direct comparisons like the ones we find in LINDEMANN occasionally. If
thus, as a slightly modified version of this doctrine states, ge- indicates merely completed action,
LINDEMANN rejects this arguing that the differences between simplex and compound were too great
to have developed solely out of the concept of completed action. But if this development lasted for
500 years, such a semantic evolution might seem quite plausible.

Doctrine 2: ge- stresses or intensifies the action of a verb

Functions like intensifying of the action or emphasis of engagement as postulated by DÜRMÜLLER/UTZ
1977 or BLOOMFIELD’s conclusion that gefaran (as opposed to faran) “appears when goal and the arri-
val point are important or unusual” 18 (1929: 101) would be very hard to prove as being in some way
systematic. And LINDEMANN is right when he asks: “What is an “intensified” going, or seeing, or stand-
ing?” (1969: 259).

Doctrine 3: ge- may convert an intransitive verb into a resultative verb that is transitive
I am not sure why LINDEMANN insists on the notion of resultativity in this matter, but that ge- com-
pounds are very often transitive when their simplexes are not can be regarded as a fact. My main
point in a proposed synthesis will be that compounding with ge- effects a change of verb class, with
a change in the positions of the verbal arguments as regular consequence, thereby moving a circum-
stantial argument into the position of the direct object. This appears to be an iconic reflection of the
fact that the formerly circumstantial argument has become central to the newly formed compound.
LINDEMANN, however, contradicts this and postulates that such verbs, usually verbs of movement,
need not necessarily be transitive, but “can still be intransitive followed by an accusative of direction
or goal” (ibid.: 261).

Doctrine 4: ge- indicates completion

This doctrine can be regarded as corresponding to the way ge- can form collectives together with
nouns, parallel to the Latin con-. Out of the meaning ‘together’ the idea of ‘being together’ with the
implication of completeness was derived. In a slightly different version, even the equivalence to Latin
cum ‘with’ was proposed, with a development of meaning to ‘fully, entirely’ and then ‘entirely to the
end’. LINDEMANN probably justly questions the semantic connection between ‘with’ and completion
(261f.). Some interesting ideas emerge when this doctrine is applied to the tense system. “The func-
tion of ge-, it said, was to indicate completed temporal action, and thereby provide verbal compounds
that would compensate for those tenses, the equivalents of Latin perfect tenses, that never devel-
oped in the Germanic languages” (ibid.). This means we are now confronted with a perfectic function
of ge-, implying that ge- compounds serve as perfect, pluperfect and future or future perfect tenses19.
For example BLOOMFIELD very frequently explains the meaning of ge- compounds as perfectic. But

18
The emphasis is mine.
19
This opinion is still widespread in modern Germanic studies, e. g. WEDDIGE claims for Middle High German that “the
prefix ge- can signify a pluperfectic understanding of the preterite (leben – gelebete) or a futuristic one of the present
(leben – geleben)” (1998: 85).
Page 7 of 27
Aspect in Old English Marc Zumstein

when one takes a closer look at his essay, the systematic value and thus the whole theory of perfec-
ticity becomes obsolete. The conclusion that sometimes, ge- compounds can convey perfectic mean-
ing, and sometimes not, accordingly that sometimes simplex verbs can do so, too, and sometimes
not, basically means nothing. LAWSON, in an essay not quoted by LINDEMANN, states the same answer
to a very similar question, namely whether the prefix gi- was a (perfectivizing) future marker in Old
High German. He writes: “We see thus far that OHG simplex verbs render both Latin present and
future, and that OHG gi- prefixed verbs also render both Latin present and future” (1965: 93). Other
works on this subject quoted by LINDEMANN come to the same result; that ge- compounds were used
to translate Latin perfects and pluperfects only at some 15 – 20%, and that “simplexes in that position
could be interpreted the same way” (LINDEMANN 1969: 262).

Doctrine 5: ge- perfectivates or expresses perfective aspect

Obviously, it is this doctrine that deserves special attention in this paper, because here we get con-
fronted with aspect. In summary, it states “that in the older Germanic dialects the simple form of the
verb, with some very few exceptions, expressed an action in its continuity whereas the compound
verb expressed an action that was cut off, or brought to an end, or completed” (ibid.). Essentially, it
was postulated that in the old Germanic languages there was an aspectual distinction similar to the
one we find in the Slavic languages, and that ge- function the same way the Slavic preverb po- does,
i. e. having been deprived of any original meaning and merely serving as a “perfectivator”, and, as a
very important feature, thereby not modifying the meaning of the simplex. As the already discussed
data shows, this theory cannot be upheld for Old English, since compounding with ge- often results
in modified meaning. The situation might have been a little different in Gothic, as will be shown.
Without anticipating too much, it also has to be said that aspect in Slavic is a compulsory grammatical
category, every verb must be either of perfective or of imperfective aspect. This is a point where for
example BLOOMFIELD gets led astray: “As a matter of fact, where OE expresses aspect, it reserves the
punctual verb (verb with prefix) for unit action, and classes repeated, habitual, and generalized acts
with the durative (uncompounded verb; more explicitly beon with present participle), exactly as does
Slavic” (1929: 92). Apart from the fact that the construction with beon plus present participle, as I
have mentioned, by no means can be regarded as a “durative” by that time, the idea that aspect can
be expressed sometimes is unacceptable.

2.3. The levels of argumentation

The analysis of Old English ge- has been identified above as being of a typological nature. This typol-
ogy does not only refer to the often-cited Slavic languages, but also to a language much closer and
much easier accessible, and that is Modern High German. First, our topic is a verbal prefix that has
retained at best very little meaning of its own, and obviously compounding with preverbs that show
only traces of own meaning is still a very productive means of word formation in ModHG (cf. the
preverbs er-, ver- or be-), but not in ModE, where only very few remnants of this old Germanic word
formation pattern have survived (cf. be-lieve versus ModHG g(e)-lauben). Second, to get a compre-
hensive impression of a (more or less) pure aspectual system, and in a second step to draw analogies
or to point out differences between the Germanic and the Slavic languages, I believe a profound
knowledge of the Slavic language system is indispensable. Therefore, scholars of English mother-
tongue suffer a particular disadvantage when doing research on this topic, and very often their ref-
erences and comparisons give the impression of a rather superficial knowledge, which naturally is

Page 8 of 27
Aspect in Old English Marc Zumstein

also true for German authors dealing with the Slavic languages20. My argumentation will therefore
involve the typological threesome ModE – ModHG – Slavic to make the situation in Old English com-
prehensible21.

3. SOME ATTEMPTS TO CLASSIFY PREDICATES

Verbs or more generally predicates act differently with regard to aspect, depending on the structure
of their so-called inherent meaning (COMRIE 1976: 41). This fact is well known to students and teachers
of English, as such a distinction plays an important role when it comes to explaining the progressive.
Sentences like I am running, I am writing or I am sleeping are well formed, while *I am knowing and
*I am reaching the top are not. A classification of predicates accounts for such “semantic aspectual
properties” (COMRIE), determining in the given case whether a certain verb can form a progressive or
not; such a classification is therefore an essential part of any discussion of aspect. In this chapter, I
will present some propositions on how to classify verbs, which will provide the necessary arguments
for my further discussion of ge- compounded verbs in Old English. I take into consideration first of all
the already “classic” Z. VENDLER with his famous “Verbs and Times” (1967), cited almost everywhere
when predicate classification is an issue. Further approaches are the one proposed by QUIRK et. al.
(1985) and the one of the Russian linguist T. V. BULYGINA22. Finally, COMRIE’s considerations will repeat
the main points and serve as a summary.

3.1. VENDLER’s classification

VENDLER uses the term “time schema” to describe “the particular way in which a verb presupposes
and involves the notion of time” (1967: 97). He also uses the possibility to form progressive tenses as
a first criterion; verbs that can form these tenses are “processes going on in time”, they “consist of
successive phases following up one another in time” (99). A second step distinguishes those predica-
tions with progressive tenses that have an internally set terminal point. So the sentences he is draw-
ing and he is drawing a circle differ in such way that if he stops drawing in the next moment, for the
first sentence it will still be true that he drew, but not necessarily for the second sentence. Here the
terminal point “has to be reached if the action is to be what it is claimed to be” (100). It gets obvious
here that a verb alone can not be classified precisely; one needs to know the surroundings, in other
words: the arguments or actants of a verb, together constituting the whole situation which is to be
classified. VENDLER proposes to call situations without a terminal point “activity terms”, and those with
a terminal point “accomplishment terms”. When analyzing those verbs that cannot form progressive
tenses, one also finds a significant difference. “Verbs like knowing and recognizing do not indicate
processes going on in time”, but “some of these verbs can be predicated only for single moments in
time, while others can be predicated for shorter or longer periods of time” (102). Those single mo-
ments in time, also frequently referred to as punctuals, are called “achievement terms”, whereas
verbs like knowing or loving are referred to as “state terms” (103).

20
A few examples to illustrate these harsh accusations: PILCH (1953: 129) is not familiar with meaning and function of
Slavic po-, COMRIE (1976: 89) translates drink up with ‘auftrinken’ (probably in analogy to ‘aufessen’).
21
I have knowledge only of one Russian scholar working on Old English aspect, namely the article of L. S. LIMAR “On the
question of the role of verbal prefixes in connection with the meaning of aspect (on the material of Old English)”. Unfor-
tunately, I had access to this article only through the quotations in LINDEMANN (1970)
22
I consider the features important for predicate classification to be to a great extent linguistic universals, so BULYGINA’s
work, which is actually a classification of Russian predicates, to me does not seem limited to the Russian language. Simi-
larly, VENDLER’s work that in principle describes English is also very popular in Russian linguistics.
Page 9 of 27
Aspect in Old English Marc Zumstein

3.2. Quirk’s classification

The classification of QUIRK et al. (1985: 201) goes into more detail, with a first differentiation between
“stative” and “dynamic” situations. Stative get subdivided according to their time reference: either
timeless “qualities” (like to be tall, to have two legs etc.) or episodical “states” (like to be happy, to
love etc.). Dynamic situations are further distinguished according to duration into “duratives” and
“punctuals”, both of which can additionally carry the features “conclusive” (implying a terminal point)
or “nonconclusive” (with no terminal point implied) and “agentive” or “nonagentive” (implying the
existence of an agent, i. e. a typically as animate perceived instigator of the action identified by the
verb). The feature of agentivity is quite irrelevant when discussing aspectual notions. QUIRK et al. fur-
ther introduce a class they call “stances”, between states and duratives, were to live, to stand, to lie,
to sit etc. belong.

3.3. BULYGINA’s classification

BULYGINA draws the most precise classification scheme, which she bases on a series of binary opposi-
tions. So as in QUIRK’s proposition, she begins by distinguishing timelessness23 and episodicity, with
timeless “qualities” (e. g. the snow is white) on one side and episodical “appearances” on the other.
These then get divided into dynamic and non-dynamic appearances, the latter being “states” (e. g.
it’s hot outside) or notions like “positions” (e. g. the picture hangs at the wall). Dynamic appearances
further split into durative “processes” and non-durative “events”. If processes additionally have the
feature perspective, they equal the accomplishment terms of VENDLER, as with perspective BULYGINA
denotes the existence of a terminal point. Processes lacking perspective equal activity terms24. Events
on the other hand are either punctuals (VENDLER’s achievement terms) or then they have a certain
extension in time. It is in the distinction between events with a temporal extension and processes with
a perspective that the specific Russian verbal system becomes evident, because here we can speak
of an aspectual difference. Both denote what VENDLER would call accomplishment terms. In the case
of events, however, the situation is regarded as a whole, i. e. perfective, whereas the process is
viewed rather from within, therefore imperfective25. With the exception of qualities and some non-
dynamic appearances, all of BULYGINA’s predicate classes can be added a feature she calls controlla-
bility, which is closely related to QUIRK’s agentivity.

3.4. COMRIE’s considerations

COMRIE does not propose a classification in the shape of a flowchart, as do the other three. He rather
limits himself to the three feature oppositions that are of crucial importance when talking about as-
pects, namely durative – punctual, telic – atelic and state – dynamic action. Punctuality, as we have
seen, is incompatible with the progressive aspect, thus *I am finding my keys is not a well-formed
sentence. Similarly, the Russian imperfective aspect can not be used to describe internal structure
with a punctual verb; an imperfective punctual verb thus always expresses iterativity. Telic situation
is the term COMRIE uses to refer to what VENDLER called the terminal point that is built into the meaning
of certain situations, whereby it may be pointed out once more that telicity refers to situations rather
than to verbs. I have, however, a slight objection to COMRIE with regards to the extension of this term.
He regards it as important that “in expressions referring to telic situations there should be both a

23
The terminology is translated from Russian.
24
Even though we are talking about completely different languages, activity and accomplishment terms still end up in the
same group.
25
To use BULYGINA’s classification outside of Russian linguistics, I would propose to join these two classes into one.
Page 10 of 27
Aspect in Old English Marc Zumstein

process leading up to the terminal point as well as the terminal point” (1976: 47), which limits telicity
to accomplishment terms. I on the contrary do not regard the process leading up to the terminal point
as essential; if a situation consists solely of the reaching of this terminal point, as it is the case with
achievement terms, then I would also call it a telic situation. When talking of telicity, the combination
with the perfective/imperfective opposition restricts the semantic range of telic verbs considerably.
If for example in Russian an accomplishment term is described in the perfective aspect, the attain-
ment of the terminal point of the situation is implied. On the contrary, the same situation in the
imperfective aspect denies the attainment, it merely states that there has been an attempt to reach
the terminal point. The verb уговорить (ugovorit’) (pf.) / уговаривать (ugovarivat’) (ipf.) exemplifies
this very nicely. The sentence он меня уговаривал, но не уговорил (on menja ugovarival, no ne
ugovoril) must be translated by lexicalizing this notion of conativity: he tried to persuade me, but he
did not succeed (literally: he persuaded (ipf.) me, but he did not persuade (pf.) me)26.
In this context, COMRIE mentions telicity as an effect of prefixation, which is highly interesting for the
subject of this paper: “In some languages, it is possible to derive verbs referring to specifically telic
situations from verbs that do not necessarily refer to telic situations, usually as a part of derivational
morphology.” To illustrate this, he lists some examples from ModHG, e. g. kämpfen ‘to fight’ versus
erkämpfen ‘to achieve by means of a fight’, or essen ‘to eat’ versus aufessen ‘to eat up’ (1976: 46).
LINDEMANN, too, has come to this conclusion: “In (the Germanic) dialects a preverb may modify the
action of a verb in such a manner as to indicate that the action tends towards a local goal27, and even
sometimes that it reaches such a goal and thereby completes the action per se, e.g., overdo, undergo,
bequeath, bypass, or uphold” (1969: 264). At this point already, I might suggest: what if Old English
ge- functioned the same way ModHG er- does in the example above, evoking not perfectivity but
telicity?
A few remarks on states and dynamic situations to end this chapter: states do not involve change, i.
e. unless something happens, a state will continue. This is why states already imply a continuous
meaning, and cannot additionally be marked as progressive aspect. For dynamic situations, on the
other hand, the distinction between continuous and perfective meaning is important. Dynamic situ-
ations involve change; to “keep going” they require a constant input of new energy. When continuous
meaning is to be expressed, the progressive aspect is used. It seems logical that the concepts of sta-
tivity and imperfectivity are closely related, COMRIE calls them a “natural combination” (ibid.: 51). The
combination of stativity with perfectivity may also occur, but usually only with reference to either the
inception or the termination of a state.

4. UNDERSTANDING AND MISUNDERSTANDING ASPECT

The aim of this chapter will be to clarify the term aspect as well as some of its fundamental concepts.
An essential part of this task is the analysis of the terminology many authors use with reference to
aspectual notions and to distinguish it mainly from the concept of aktionsart. Some authors treat
aspect and aktionsart either as synonyms (PILCH 1953) or just use both terms, presumably in different
meanings, but without explicit definitions (LINDEMANN 1969). The concept of aktionsart will be the
subject of the following chapter, so here a brief definition may suffice. While aspect describes “dif-
ferent ways of viewing the internal temporal constituency of a situation” (COMRIE 1976: 3), thus ex-
plicitly not changing the lexical meaning of the predicate, aktionsarten denote a temporal or modal

26
The example is taken from COMRIE 1976: 48.
27
The emphasis is mine.
Page 11 of 27
Aspect in Old English Marc Zumstein

modification of the meaning expressed by the simplex. This modification of meaning is evoked
through derivational morphology patterns, which carry a more or less systematic character in the
given language28.
Without again reiterating definitions of aspectual oppositions that I have already discussed, I will at
this point try to establish a relationship between the two oppositions imperfective/perfective and
progressive/nonprogressive. While the perfective aspect is largely dominated by the notion of total-
ity, meaning that it views a situation as a single whole, the semantic range of the imperfective aspect
is wider29. Thus imperfective situations, not regarded as a total, can either be habitual or continuous.
According to the distinction pointed out above, the semantic feature of stativity is decisive whether
a continuous situation can form a progressive or not (COMRIE 1976: 25). When applying this to ModE,
we see that the progressive aspect covers only a small part of an imperfective aspect, whereas the
English nonprogressive is used as well in perfective situations as in the imperfective habitual and
continuous nonprogressive situations. RAITH (1951: 13, 29) commits this error to use the terms im-
perfective/perfective with reference to the ModE verbal system, he claims I am writing a letter to be
an imperfective, accordingly I write a letter every day to be a perfective aspect, used to express “time-
less situations”, which is, in my eyes, a contradictio in adjecto30. In does not even make sense in the
light of RAITH’s own explication of aspect, which can be summarized as follows: RAITH describes aspect
as “relation of time direction”. He states that there are two possibilities to reflect time in language,
one assuming a fixed time scale (or “time line”) with “present point” on it, moving from the past into
the future, the other with a fixed present point on a time scale moving from the future into the past.
These two different viewings of time he then identifies with imperfective and perfective aspect re-
spectively (ibid.: 28). Apart from the fact that I really did not grasp the point RAITH is trying to make, I
consider this to be confusing rather than explanatory.
Another frequent terminological error is the use of durative as a synonym for imperfective and punc-
tual (or momentary) for perfective (for example RAITH 1951: 21 or BLOOMFIELD 1929: 92). A simple
example may illustrate this; given the almost prototypical aspectual system of Russian, it is taken
from this language. In the sentence он постоял там час (on postojal tam čas) ‘he stood there for an
hour’, the perfective aspect indicates that the situation of him standing there for an hour is viewed
as a whole. As the duration for an hour is even lexically expressed, the situation obviously is durative
and thus non-punctual.
If aspectual allusions are involved, ge- prefixed compounds are generally claimed to form or to evoke
perfective aspect or at least perfective meaning (cf. LINDEMANN’s doctrine 5). Therefore, it makes
sense to give more detailed definition of the perfective, thereby underlining the features that might
wrongly be identified with it, and to compare these features with the perfect in ModE, which is some-
times also called “perfective aspect”. Perfectivity does not distinct the various separate phases that
make up a situation (COMRIE 1976: 16). This does not imply that it necessarily indicates short duration
(any duration is possible, as long as the situation is still viewed in totality) nor limited duration, as
opposed to imperfective unlimited duration (between il régna trente ans and il régnait trente ans, the
difference is not one of limit, but purely one of aspect). To regard a situation as a whole does not
exclude internal complexity of the given situation; the perfective verb in он позапирал все двери
(on pozapiral vse dveri) ‘he closed all the doors one after the other’ exemplifies this. Also, a perfective

28
This definition is taken from Metzler Lexikon Sprache.
29
This is also the reason why the perfective aspect is usually referred to as the marked member of the aspectual opposi-
tion.
30
We might slightly relativize COMRIE’s relation between progressive and imperfective (and question RAITH’s terminology
once again) when we consider the past perfect progressive, which expresses durativity as well as perfectivity (the following
paragraph aims to show that these two notions do not contradict each other).
Page 12 of 27
Aspect in Old English Marc Zumstein

situation need not necessarily be completed, this term laying too much emphasis on the termination
of the situation. The perfective describes a situation as complete, with beginning, middle and end.
Similarly, resultativity can at best be claimed for perfective and thus successful accomplishment
terms.
On the other hand, the general definition of the perfect as “the continuing relevance of a previous
situation” (COMRIE 1976: 56) at least inherently contradicts the basic notions of perfectivity, com-
pleteness and totality. From the four possible meanings the perfect can express (COMRIE 1976: 56 –
60), only the perfect of result bears some similarity to the perfective aspect, the result being a pre-
ceding situation presented in totality as a basis for a new situation (cf. he has arrived ‘he is here’).
Contrary to this, the experiential perfect (cf. I have already been to London) translates as an imper-
fective into Russian, because the mere information where some situation has ever taken place in the
past does not require the notion of totality, as in reference to a concrete situation. Also, the perfect
of persistent situation explicitly excludes totality, and the perfect of recent past (cf. I have just called
him) is a pure matter of tense.
We should not end this chapter on aspect without having mentioned the way MITCHELL (1985: 366)
qualifies the many confusing attempts to detect traces of aspectuality in Old English, which, as has
been discussed in the last chapter, rather referred to the inherent semantics of different predicate
classes: “… there is no more reason why we should establish special categories of verbs according as
they denote completeness or incompleteness, beginning or end, etc., than why we should establish
categories according as they denote something hard or soft, or something pleasant or unpleasant (he
patted her cheek: soft aspect; he smacked her face: hard aspect!). Once we leave the ground of in-
flectional form, we are lost in a chaos of infinite possibilities of distinction.”31. This issue about the
inflectional form as distinctive marker will also be of great importance in the following chapter.

5. TWO COMPETING IDEAS ABOUT AKTIONSARTEN

In the works of the authors consulted for this paper, the term aktionsart is not used homogeneously;
we can rather distinguish two main directions of understanding, one of which I would place closer to
those authors concerned with less inflecting languages (such as the Germanic languages) and the
other one aiming more at the description of languages with a well/developed inflectional system
(such as the Slavic languages). To describe these two approaches, COMRIE (1976: 6f) paraphrases as-
pect as the “grammaticalisation of semantic distinctions”, and aktionsart either as the “lexicalisation
of such distinctions, irrespective of how these distinctions are lexicalized” (which is, as he states him-
self, very close to the notion of inherent meaning), or as “lexicalisation of the distinction, provided
that the lexicalisation happens by means of derivational morphology”. Unfortunately, elements of
these two definitions can also be found mixed up.
As I have mentioned already in the preceding chapter, I consider the morphologically expressed ak-
tionsarten to be the correct representation of this language category, whereas for the other possibil-
ity I prefer the term predicate class. “Purely semantically supported definitions of aktionsarten are
problematic, as there are no language-internal structural facts corresponding to them; they are ra-
ther semantic determinations specific for a given language, based on the classification of the verbs
of a language into semantic groups.”32 An important point to be added to the above definition of
aktionsart is the fact that all verbs specified in terms of aktionsart must have a simplex counterpart,

31
MITCHELL here quotes BODELSEN (EStudies 32, 1951, pp. 259 – 260)
32
This and the following are based on the explication given in Metzler Lexikon Sprache.
Page 13 of 27
Aspect in Old English Marc Zumstein

the basic lexical meaning of which they modify. Further this semantic modification is systematic in
such a way that the same type of morphological derivation always modifies a given meaning in the
same way, e. g. the Russian preverb za-, denoting the beginning of a situation: лаять (lajat’) ‘to bark’
– залаять (zalajat’) ‘to start barking’, говорить (govorit’) ‘talk’ – заговорить (zagovorit’) ‘to start talk-
ing’. In German, such aktionsarten did not develop systematically, solely a diminutive – iterative ak-
tionsart, expressed through the infix -l- and sometimes a change of vocal: husten ‘to cough’ – hüsteln
‘to cough a little, repeatedly’.
A. V. ISAČENKO, who has written the classic article about aktionsarten in Russian and who understands
it according to the above definition, begins his discussion by criticizing the “wrong” use with reference
to Germanic languages. Thus find is regarded as the “successful completion” of search, ModHG töten
‘to kill’ is claimed to be a “causative” to tot ‘dead’ and verbs like breathe would be “natural iteratives”,
consisting of taking many breaths (1962: 385). But these verbs are all simplexes with a primary, and
not derived, meaning. Any derivational relationship is in most cases historic and not productive, usu-
ally not even understood as a derivation anymore, e. g. schneiden ‘to cut’ and “derived” from it as
“intensive aktionsart” schnitzen ‘carve’33. Russian aktionsarten, on the other hand, are always marked
by one semantic and two formal characteristics, a formal marker (usually a prefix or a suffix), a mod-
ification of the meaning of the simplex in a way that focuses on details within the situation, and the
lack of an aspectual counterpart to the so formed aktionsart. Therefore, it is quite a regular principle
in Russian aspect theory that only telic verbs can form so-called aspectual pairs, while non-telic verbs
can form series of aktionsarten (in either aspect) focusing the attention on parts within the described
situation. Contrary to aspectual system, where every verb has to be part of, there is no such thing as
an opposition with regards to aktionsarten; there are only a limited number of verbs that can form
them at all. A very important fact mentioned by ISAČENKO is that the theory that the formation of
aktionsarten was a part of the development of the modern aspectual system. This can be seen in Old
Church Slavonic, where the aspects were still emerging out of a tense system similar to the one in
Ancient Greek.
But now we have to take a closer look at the way this term is used and misused by the authors con-
sulted; RAITH 1951, STREITBERG 1920, WEDDIGE 1998 and obviously with them many more writing on
Germanic languages define aktionsarten as “the manner in which the action described by the verb
proceeds”, as such a rather precise description of inherent meaning. Confusion begins, when RAITH
(1951: 21 – 24) in this sense starts to talk about perfective and imperfective aktionsarten, forming
pairs as search (ipf.) and find (pf.) or schweigen (ipf.) ‘be quiet’ and verstummen (pf.) ‘become quiet’.
First of all, the pairing of search and find seems highly questionable, as these are two entirely different
verbs, linked solely through “world knowledge”. But also climb and climb to the top, according to
RAITH imperfective and perfective, or erschlagen ‘kill’ and verwelken ‘whither away’, identified as per-
fective and resultative respectively by WEDDIGE (1998: 85)34 are not very convincing. It must be made
very clear, once again, that the concepts described are not untrue, but they confuse morphologically
expressed aspectual differences (of which aktionsarten also form a part) with semantically inherent
meanings of time reference35.
A good example of the effect of such confusion is PILCH 1953. The author completely misunderstands
the term aktionsarten in the work he attempts to criticize, taking them for formal and semantic equiv-
alents to the Slavic aspects, whereas they obviously denote predicate classes. PILCH claims that

33
From the synchronic point of view, also the relationship between these two meanings is not comprehensible anymore.
34
WEDDIGE, however, also mentions lächeln ‘smile’ as diminutive aktionsart to lachen ‘laugh’, which again is a morpholog-
ically expressed aktionsart according to our definition.
35
Not only the reader, but in the case of RAITH 1951, to my opinion, also the author gets lost in his explications.
Page 14 of 27
Aspect in Old English Marc Zumstein

STREITBERG took Gothic sitan/gasitan for an equivalent of Slavic sěděti/sěsti ‘sit/sit down’, thereby mis-
takenly regarding both as aspectual pairs. In my reading at least of his Gotisches Elementarbuch,
STREITBERG’s theory works differently. Due to his lack of modern linguistic terminology, one has to use
the way he analyses his examples to find out what he meant; using his terminology on the other hand
can hardly lead anywhere. Let us look at an example: STREITBERG (1920: 195f.) states that by means of
the perfective aktionsart “a situation is marked with respect to the moment of its completion”. To
illustrate this, he also uses the ModHG ersteigen ‘climb to the top’, implying the relation to the sim-
plex steigen ‘climb’. His explication is that ersteigen denotes “the action of climbing only with regards
to the moment of termination”, which semantically meets the point, thereby fitting into his system
of aktionsarten as predicate classes. In VENDLER’s words, steigen is an “activity term” and ersteigen an
“achievement term” (cf. VENDLER’s own example to reach the top; 1967: 102f.). This can be shown
graphically as follows:

steigen: … →→→→→→→→→→ …
ersteigen: (→→→) •

Thus in this particular case, the ModHG compound with the preverb er- focuses on a possible result
of the action described by the simplex. Through this terminal point that has been added to the com-
pounds meaning, it changes the predicate class and becomes telic, while the simplex alone does not
contain any implication about a goal or a result. This notion is of course not explicitly mentioned by
STREITBERG, but I believe this was more or less what he meant. To return to the original question about
the relation between sitan/gasitan and sěděti/sěsti respectively, the semantic opposition obviously
appears to be of a very similar kind, as the graphic representation shows:

sitan - sěděti ‘sit’ … →→→→→→→→→→ …

gasitan - sěsti ‘sit down’ • (→→→)

Unfortunately, I did not have STREITBERG’s main work on this topic (Perfective und Imperfective Aktion-
sart im Germanischen, PBB 15, 70-177) at my disposition, and therefore I can only refer to his already
quoted book. But as I have tried to show, on the background of his illustrations, I do not believe
STREITBERG ever intended to postulate an aspectual pair sitan/gasitan parallel to the Slavic sěsti/saditi
sę (which is, in my opinion and contrary to PILCH36, the corresponding aspectual pair for Old Church
Slavonic). His issue was probably rather to show a changing of the predicate class, which might even
be regarded as a kind of an ingressive aktionsart in this particular case. In ModHG, the preverb er-
can serve to mark the reaching of a goal, e. g. the often-cited erkämpfen ‘attain through fighting’ (e.
g. COMRIE 1976: 46) or erwürgen ‘strangle (with lethal consequences)’, as opposed to würgen ‘strangle
(with lethal consequences)’, thereby usually effecting changes in the actant structure of a predicate.
But even though semantically this looks like a valid resultative aktionsart, evoked through compound-
ing with er-, still this cannot be claimed to be systematic in ModHG, as there are also many instances
where this preverb results in a change of meaning, like schwindeln ‘lie’ and erschwindeln ‘attain
through lying’. With others the meaning is not affected, like forschen/erforschen ‘investigate, do re-
search’ (and not necessarily: ‘attain a result of one’s research’). So differences in the semantic

36
The pair he mentions, sěsti/sědati, appears to be taken from Old Czech.
Page 15 of 27
Aspect in Old English Marc Zumstein

structure of this pair are mainly due to different actant structures and very often stylistic nuances
due to differing contexts or fields of use.

6. FUNCTIONS AND EFFECTS OF PREFIXATION

6.1. Germanic languages

The question about meaning and function of the Old English preverb ge- can be concentrated to the
general question about the effect of more or less desemantized prefixes. Now this is an appearance
found not only in the Germanic languages (not in ModE, but still in ModHG), but also in the Slavic
languages, which again has often led to the postulation of more or less justified analogies. RAITH
(1951: 24f.), basically repeating STREITBERG (1920: 195), heads in the right direction when he assumes
that in the Indo-European languages, compounded verbs are usually perfective and simplex verbs
imperfective. Also his assertion that the more a preverb loses its meaning, the more it functions as a
pure perfectivator, even though the terminology has to be understood cum grano salis. But even
COMRIE (1951: 90f.) writes: “Thus in the Germanic languages there are some prefixed verbs that can
have perfective meaning”, where I would not use the term perfective, but he continues that “there is
no systematic pairing of forms with perfective and imperfective meaning, and in general no way of
deriving forms with imperfective meaning from verbs with perfective meaning. (It is possible that the
prefix ga-, as in Gothic, originally denoted, or came to denote in some Germanic languages perfective
meaning, but outside of Gothic37 there is little direct evidence of such a situation …)”. In his theory
about the emerging of aspectual oppositions, it is the prefixation with a meaningless38 prefix that
plays a crucial role: “The criteria that go most towards making a systematic set of aspectual opposi-
tions are: the presence of an otherwise semantically empty perfectivizing prefix (such as Gothic ga-,
Hungarian meg-, Georgian da-, Lithuanian pa-, Slavonic po-, or other prefixes with more restricted
sets of verbs); the possibility of forming imperfectives from verbs where the prefix changes meaning
other than just aspect (…)” (1951: 94). I doubt that in these cases the term perfective can already be
used, but these prefixes definitely added an internal terminal point to the meaning of the verb, which
could then serve as a basis for the development of an aspectual system, which it did in the Slavic
languages, but not in Germanic. COMRIE’s further statements (1951: 93f.) become somewhat hum-
bler: “The languages examined that have prefixes or verbal particles with, at least sometimes, aspec-
tual (perfective) significance, can be arranged along the following scale according to the extent to
which they have a fully developed system of oppositions between Perfective and Imperfective, start-
ing with those languages with the least fully developed system: English and German, Hungarian, Bal-
tic, Georgian, Slavonic.” How exactly aspects develop out of prefixation, COMRIE puts as follows: “The
addition of a prefix to a simple verb normally results in a restriction of the meaning of that verb, and
one way in which such a restriction can be interpreted is as a restriction to a single unified complete
action; this is by no means a necessary restriction, as is shown by those languages where verbal pre-
fixes do not normally have aspectual significance, though particularly in the presence of semantically
neutral aspectual prefixes and of processes of imperfectivisation, this can lead to the development
of an aspectual system relying primarily on prefixation as a means of perfectivisation.” For the Ger-
manic languages, this process ended with that restriction of the meaning of the verb in question that
I claim to be principally and at least originally an addition of the inherent terminal point. As the

37
This supports my assumption that some significant changes in the system of the Germanic languages could have taken
place between the time our Gothic evidence originates from and the time we get sources of the other Old Germanic
languages.
38
STREITBERG and RAITH call it “colourless”.
Page 16 of 27
Aspect in Old English Marc Zumstein

situation in ModHG shows, that original meaning was very likely to develop a semantic of its own in
the course of evolution. From this point of view, MITCHELL (1985: 367) and RAITH (1951: 25) are also
correct in stating that simplex and compound verbs describe two different situations and are thus to
be classified as different verbs, “they are a matter for the lexicographer” as MITCHELL puts it. Aspectual
pairs on the contrary describe exactly the same situation and carry identical lexical meaning39.

6.2. The Slavic prefix по-/po-

In the Slavic languages, with Russian as its most eminent representative, the existence of aspect as a
verbal category results in a grammatically more evident functioning of prefixation. ISAČENKO writes
that “the combination of a simplex with a verbal prefix in all cases results in a determination of the
verb in question. (…) This narrowing of the original meaning evoked through the addition of a verbal
prefix – i. e. this determination of the verb – in all Slavic languages coincides with the general meaning
of the perfective aspect” (1962: 356). As a consequence, prefixation regularly transforms an imper-
fective simplex into a perfective compound. Of course, different prefixes add different meanings to
the simplex, e. g. the prefix вы- (vy-) ‘out-’: и(д)ти (iti) (ipf.) ‘to go’ – выйти (vyjti) (pf.) ‘to go out’,
дать (dat’) (ipf.) ‘to give’ – выдать (vydat’) (pf.) ‘to hand out’. Sometimes, a prefix gave up its meaning
to form an aspectual pair, or the prefix appeared desemantized if it “simply reiterated some inherent
semantic feature of the verb” (COMRIE 1976: 89), as in the aspectual pair читать/прочитать
(čitat’/pročitat’) ‘to read’. The prefix про- (pro-) means ‘through’, so прочитать (pročitat’) actually
means ‘read through’.
The crucial point for our discussion, however, is the preverb po-, as it appears quite frequently to
illustrate supposed or denied parallels to ge-, but even po- has a character of its own. PILCH (1953:
129) cites a work of STREITBERG, where obviously po- and ge- were claimed to be almost identical in
function – meaningless elements merely adding the notion of perfectivity to a given simplex. PILCH
challenges the idea that “the material content of по-/po- is reduced to zero” with a sentence from a
canonical Russian grammatician40, stating that on the contrary po- contained a wide range of mean-
ings with a remarkable diversity. This is, of course, true, but not presented correctly, as I intend to
show. Indeed, when used to form aktionsarten, po- can express four different meanings, namely in-
gressive (чувствовать (čuvstvovat’) ‘to feel’ - почувствовать (počuvstvovat’) ‘to start feeling’), limita-
tive (постоять (stojat’) ‘to stand’ – постоять (postojat’) ‘to stand for a certain while), attenuative
(отдохнуть (otdoxnut’) ‘to rest’ – поотдохнуть (pootdoxnut’) ‘to rest a little bit’) and distributive
(запирать двери (zapirat’ dveri) ‘close the doors’ – позапирать двери (pozapirat’ dveri) ‘close all the
doors, one after the other’). But as has been said, only a limited number of verbs can form aktion-
sarten, so when GUIRAUD-WEBER examined 1000 verbs prefixed with po-, she found 692 of them to be
completely meaningless. She concludes that “po- is probably the most often desemantized preverb
in modern Russian: in the vast majority of cases its only function is to perfectivize an imperfective
simplex without modifying its lexical meaning” (1993: 57).

39
For further discussion on this topic, cf. ISAČENKO 1962: 355f.
40
V. V. VINOGRADOV (1947): Russkij Jazyk (‘The Russian Language’), Moscow
Page 17 of 27
Aspect in Old English Marc Zumstein

7. A NEW THEORY ABOUT GE-

7.1. Some predecessors

Before I start to analyze LINDEMANN’s explication of the meaning of ge- as the last “highlight” of this
paper, it seems reasonable, on the background of the point discussed above, to take a quick look at
some earlier attempts to do so. RAITH (1951: III) states that “perfectivization with the prefix ge- is not
connected with the distinction of aspects”, i. e. he uses aspect only for the progressive – nonprogres-
sive opposition, whereas ge- expresses aktionsart. My assumption that he actually refers to predicate
classes is supported by RAITH’s claiming that simplex and ge- compound form two entirely different
verbs, and also the narrowing of time reference is mentioned: “Obviously compounding destroys the
notion of duration, inherent in the simplex, and summarizes the process of time in one point” (ibid.:
24). The fact that there are simplexes that already carry the inherent notion of telicity is among others
pointed out by STREITBERG, who notices that although most Gothic simplex verbs are “durative”, there
are some that never combine with ga-, not even in the present participle. This formal criterion, in
addition to their lexical meaning, marks them as “perfective aktionsart”, i. e. as achievement terms.
Examples are fraþjan ‘to understand’, finþan ‘to recognize’, giban ‘to give’ or qiman ‘to come’ (1920:
198). Very interesting is also his remark that duginnan ‘to begin’ cannot combine with achievement
terms, cf. *he started finding his book. STREITBERG, however, suggests that generally ga- compounds
could not follow duginnan, implying that they all need to be regarded as achievement terms. His evi-
dence supports this view, but the forming of accomplishment terms by prefixation is at least theoret-
ically also possible41. As a matter of fact, it is beyond my possibilities in this paper to prove my theses,
but I will try to exemplify them at the case of the Gothic verb hailjan, which is glossed with its ModE
cognate ‘to heal’ (cf. Old English hælan). I see here a possible analogy to the Russian лечить (lečit’),
which is also usually translated by ‘to heal’, but which in fact only means ‘to treat (medically), to cure’,
with no reference to success. Only in combination with the already mentioned prefix вы- (vy-) ‘out-’,
the meaning is modified as ‘to attain the goal of treating’, i.e. ‘to heal’. So my thesis reads that such
a semantic relationship existed also in Germanic.
BLOOMFIELD’s contribution does not bring much light into the dark. He analyzes the use of eight Old
English verbs and there ge- compounds in the works of Alfred the Great, but in the end, he fails to
provide the reader with a coherent theory about use and function of simplexes and compounds. He
rather treats every verb as a single phenomenon for which he lists specific patterns of use, but with-
out gaining any conclusions about a general function of ge-. He operates with notions such as differ-
ent categories of objects, technical terms, abstract or concrete reference or even emphatic notions,
and in line with these sometimes even the term aspect appears.
PILCH (1953) divides appearances into indivisible and divisible ge-, the second referring to the prefix
as a part of inflectional morphological in the preterite participle. Indivisible ge- is further subdivided
into three formal groups, one with verbs that rarely or never form ge- compounds (his examples are
verbs like beon, cuman or findan), one with verbs that only or nearly only appear with the preverb
(verbs like gebæran, gefredan or getimian) and last but not least those that show both simplex and
compound verbs. These again PILCH parts into pairs with and pairs without evident differences in
meaning, thus partly repeating the doctrine about the meaninglessness of ge- (ibid.: 133). He gives
the following two meanings and functions of the preverb: ge- acts as an “aiming or reinforcement of
the thrust onto the object”, which is how he accounts for the transitivity of most of the compounds;

41
An example from ModHG would be sehen ‘to see’ versus besehen ‘to take a look at something’, whereby the compound
combines with adverbials of duration.
Page 18 of 27
Aspect in Old English Marc Zumstein

and further that ge- underlines the involvement of the subject in a given situation42 (ibid.: 135f.).
PILCH also states that verbs of perception and verbs of happening tend to generally prefer the com-
pounds. All those cases where his assumptions do not work he claims ge- compounds to be “old
frozen forms inherited from West Germanic times”, where semantic shifts in the course of develop-
ment obscured original meaning. Unfortunately, he does not propose any theories for those old
meanings that have been obscured.

7.2. LINDEMANN’S new theory

As I have said in my initial remarks on ge-, a language element completely deprived of any meaning
or function whatsoever, be they of phonemic, morphemic or syntactic nature, runs contrary to any
logic. LINDEMANN, who has been working on the prefix ge- obviously for many years, is determined to
explain it by tracking down its original lexical meaning, because “to assume that the reflexes of *ga-
were without concrete meaning runs contrary to logic and empirical evidence” (1970: 23). And for
him who understands aspect only in such way that a meaningless marker turns imperfective sim-
plexes into perfective compounds, the revelation of such a meaning would make the “theory of as-
pect fly out of the window” (ibid.: 22). A first issue claiming that the meaning was ‘with’ or ‘together’
LINDEMANN rejects, because “during all of the three hundred years that have been devoted to discus-
sions of the morpheme no one has ever proved that ge- or any of its cognates meant ‘with’, ‘to-
gether’. Evidence for it is absolutely nil” (ibid.: 24). This assumption was based on the identification
with Latin com-, which itself was supposed to be linked to the preposition cum ‘with’. LINDEMANN
argues that this semantic link between a preposition and a preverb can by no means be taken as
given, and on the other hand also statistic evidence is against an identification of ge- and com-, as for
example in the Anglo-Saxon Gospels ge- translates the Latin preverb ad- just as often (ibid.). Thereby
he still claims com- and ge- to express very similar meanings, implying thus that also com- and ad-
are quite close in meaning and that there has been a transition of preference from com- in earlier
periods of Latin towards ad- in the later periods. This close semantic vicinity of preverbs had a parallel
in OHG, where com- was glossed with pi- (cf. ModHG be-) almost as frequently as with gi-43. LINDEMANN
concludes that “OE preverbal ge- would have to “mean” that the action expressed by any verb to
which it is prefixed is directed toward some thing or in a direction forward and outward” (ibid.: 37).
He bases this result on semantic evidence gathered by scholars in three centuries, namely that ge-
corresponded primarily to such Latin proverbs as ad-, com-, in- and ex-, that ge- carried similar no-
tions as the German prepositions an, zu, hin(weg), über – hin, auf and the preverbs be-, er-. Addition-
ally, his own contribution consists in a comparison of the use of preverbs in the Gospel translations
of the Old English dialects West Saxon, Kentish, Mercian and Northumbrian. As his evidence proves,
ge- compounds in one Old English dialect correspond to compounds in a-, be-, for-, forþ-, on- and to-
in the other OE dialects (ibid.: 36). One example may be cited, the translation of Matthew 25, 18,
which in Latin reads: et abscondit pecuniam domini sui ‘and buried the money of his master’. West
Saxon translates this as and behydde hys hlafurdes feoh, whereas Mercian uses the verb ahydde and
Northumbrian finally gehydde. I understand the main argument to be interchangeability of preverbs,
a phenomenon also apparent in ModHG, where prefix shifts were common as late as in the 18th cen-
tury, e. g. between be- and ge-; modern geschehen ‘happened’ as opposed to earlier beschehen. The
other hypothesis of LINDEMANN, claiming that the spatial meaning, which admittedly is very likely to
have been there originally, was still relevant in the times of Old English, seems rather questionable.

42
These are the doctrines 2 and 3 according to LINDEMANN 1969.
43
Already Jacob GRIMM has pointed out in his Deutsche Grammatik of 1878 that the meaning of ge- was very closely
related to that of ModHG preverbs such as be- or er-.
Page 19 of 27
Aspect in Old English Marc Zumstein

But this is how LINDEMANN accounts for the often-evoked transitivity, which he claims not to be one,
and “the accusative next to it must therefore be understood as an accusative of direction” (ibid.: 65).
The already cited sentence from Alfred’s Orosius must therefore be slightly modified: þa siglde he
þanon suþryhte be londe swa swa he mehte on fif dagum gesiglan ‘then he sailed from that place
straight south along the land as far as he was able to sail out (to) in five days’44. Apart from this rather
inelegant translation, I think the feature of telicity would bring us further than that of direction, nota
bene denoting the same phenomenon. The point that can be reached after a cruise of five days is at
the same time the point of reference of the directional aim claimed by LINDEMANN to be added through
compounding with ge- and the inherent terminal point added to the semantics of siglan. With regards
to the question of transitivity, I might again use the analogy to ModHG to exemplify this problem. As
has been pointed out more than once, prefixation of a simplex leads to a determination or a narrow-
ing of its meaning, often focusing on a specific part of the situation, and, as I have stated when dis-
cussing doctrine 3, moving this part to a more prominent position from the point of view of sentence
semantics.

7.3. Why did ge- disappear?

Despite of the fact that ge- had a very eminent presence in the lexicon of Old English, it has nonethe-
less disappeared in the course of the Middle English period, with only scarce traces surviving into
ModE. The etymological dictionary of Kluge lists the following examples as remnants of Old English
ge-: i- in handicraft, y- in everywhere and e- in enough; PILCH (1955: 46) additionally mentions afford
(< ME i-forþien < OE ge-forþian). A different situation we are presented in ModHG, where ge- has
become a morphological marker of the preterite participle45. Probably such an evolution was also
beginning to happen in Old English (which accounts for ge- in the preterite participle also of simplex
verbs), but the Norman invasion of 1066 AD with its massive impact on the English language caused
matters to develop differently.
PILCH (1955) claims that ge- was lost due to phonological reasons. Between the 10th and the 12th
century ge- developed into i-, which is plausible given the fact that already in the Old English times it
had probably been pronounced as /je/. PILCH now argues that words in i- were often preceded by
words ending in the same or similar vowels, resulting in a hiatus. These were then contracted, so that
new simplexes emerged from the original compounds: hi iseoþ > hiseoþ > hi seoþ (ibid.: 39f.). Accord-
ing to PILCH, this leads to the emerging of a sentence-phonetic function, meaning that forms in i- were
used when the preceding vowel did not end in a vowel. In a next step, the unprefixed forms expanded
their range, whereby i- loses its semantic value and becomes grammatical (as in ModHG). In a last
step, analogy causes i- to vanish in all verbal forms (ibid.: 47). In the late Middle English period, a
partly revival of i- can be witnessed, but now it functions solely to mark an archaic style, and accord-
ingly appears also where it should never be from a semantic and grammatical point of view. LINDE-
MANN regards matters differently: “The loss of the morpheme was due not to phonological weaken-
ing, as Herbert PILCH suggests, but probably to the fact that during the Middle English period the
whole OE preverbal system broke down and was supplanted by verb-adverb, verb-preposition com-
binations” (1970: 65). I also believe PILCH’s issues to have been of lesser importance, that it was rather
the development of the English language towards an isolating-analytic language that has made deri-
vational patterns like prefixation with ge-, be- and other preverbs widely superfluous, as for example
the change of word class can often be attained trough simple conversion, cf. the following sentence

44
The translation is from LINDEMANN.
45
But, interestingly, not in those cases where there is another desemantized preverb: zurückgehen ‘go back’ – zurückge-
gangen/*zurückgangen but begehen ‘walk on, celebrate’ - *begegangen/begangen.
Page 20 of 27
Aspect in Old English Marc Zumstein

in ModE and ModHG: he wants his girlfriend to mother him versus er möchte, dass seine Freundin ihn
bemuttert. Very few such compounds survived into ModE, if they did, as behead, they had become
lexicalized.

8. THE EXPANDED FORM IN OLD ENGLISH

I cannot conclude a paper dealing with Old English aspect without having at least mentioned the
expanded form as it presented itself in these early days of English. Evidently already by that time
some kind of periphrastic verbal forms were in use, as the often-quoted example of Chronicles 755
shows: ond hie ealle on þone cyning wærun feohtende, oþþæt hie hine ofslægenne hæfdon ‘and they
were all fighting against the king until they had slain him’, where the tenses even match ModE pat-
terns perfectly. But synchronically they may still only be regarded as a stylistic variant, as is pointed
out in most grammars of Old English. Their appearance is usually said to be connected translation of
similar Latin forms, according to some authors Latin was simply copied by Old English, others claim
Latin only to have triggered a latent feature or tendency, basically available in Old English. As RAITH
(1951: 111) mentions, in Late Latin such periphrastic tenses become more and more popular, it
seems that a feature of Latin everyday speech made its way into the literary language. MITCHELL (1985:
279) sees the Old English expanded form as a “blend of three syntactic constructions”, firstly the verb
to be plus a predicative adjective as in hie wæron blissiende, secondly to be plus an appositive parti-
ciple as in he wæs on temple lærende his discipulas and finally to be plus an agent noun in -end as in
he wæs ehtend cristenra manna, where ehtend can be glossed as ‘prosecutor’. MITCHELL also notes
three factors that he claims to have assisted in forming the periphrastic tenses, apart from the already
mentioned impact of Latin participle constructions these are a “general trend in OE to build up a so
called analytic form system using free instead of bound forms” and surprisingly a “tendency to re-
establish the system of aspectual relations, which had broken down in primitive Old English”46. With
regards to the functioning, RAITH (1951: 98f.) analyzes the Old English periphrastic tenses with the
features of ModE progressive tenses, that on-going situations as well as background situations only
stand in the progressive, whereas timeless situations can only be presented in the nonprogressive. In
Old English, however, all these three principles are “violated”. Most authors also silently accept the
implication that these Old English tendencies later developed into the modern progressive tenses,
only BARBER (1993: 162f.) presumes that these tenses arose only in the Middle English period, thus
not descended from the Old English usage. His arguments are morphological: “It is more likely that
they arose from ME sentences like he was areading, where areading has developed from on reading
(…). Originally this reading was not part of the verb, but was a noun (OE ræding), meaning ‘the act of
reading’.”

46
Such “primitive aspectual relations” would need to have some kind of reflection in Gothic. This is not the case, and it is
also astonishing that MITCHELL accepts this notion, because later he categorically rejects any connection between aspect
and the Old English verb.
Page 21 of 27
Aspect in Old English Marc Zumstein

9. SUMMARY

“There can be no doubt that it is possible to give a coherent account of the OE verb without using
the term ‘aspect’” (MITCHELL 1985: 364). This is, stated very briefly, the summary of the present paper
on aspect in Old English. What this term stands for in Modern English, the progressive – nonprogres-
sive opposition, is not or not yet a relevant category in Old English. Periphrastic forms exist, but they
are purely stylistic variations. The main part of the paper dealt with another aspectual opposition,
imperfective and perfective. Its issue was to clarify whether there was any such opposition in Old
English, and if not, why aspect keeps getting mentioned in the works of scholars. In close connection
with this or rather the main point in this discussion was the meaning and function of the Old English
preverb ge-, which has often been claimed to form a perfective aktionsart when added to a simplex.
I will present the results of my work in the form of theses:
1. There was no such thing as an aspectual opposition in Old English, neither progres-
sive/nonprogressive nor imperfective/perfective. An aspectual opposition must be sys-
tematic, dividing all possible verb forms into either perfective or imperfective; thus one
part of the opposition can also always be defined from the other part. In Old English, a
perfective can not be postulated, as there is no imperfective aspect in first place. The
Russian linguist LIMAR47 puts this as follows: “In Old English ‘essentially there was no form
for the imperfective aspect’, and since there was no imperfective aspect, that is, one of
a correlative pair was absent, there obviously could be no system of aspects, since the
system is built on correlatives.”
2. There were no aktionsarten in the precise definition of the term, meaning morphologi-
cally derived forms of simplex verbs that alter the meaning of these verbs systematically,
bringing into focus a part of the described situation, usually temporally or modally.
3. The term aktionsart as it is commonly used by scholars writing about the older Germanic
languages describes a concept I would rather refer to as predicate classes, their most
important distinctions being punctual – durative, stative – dynamic and telic – atelic.
Prefixation has been shown to determine the meaning of a simplex verb, thereby often effecting a
change of predicate class and of the actant structure. If the added prefix is more or less deseman-
tized, the meaning of the resulting compound is likely to be a narrowed meaning of the simplex. The
determination evoked by prefixation has been identified to be an initial phase towards the forming
of a system of morphological aktionsarten, which then can serve as a basis for the development of
an aspectual system.
4. Prefixation with ge- in Old English probably in first place effected a change in verb class,
usually adding the notion of telicity or even punctuality. The inherent semantic nature of
these notions was often slightly misunderstood and interpreted as perfectivity.
5. In this quality, ge- was interchangeable in Old English and its dialects with a series of other
preverbs that are similarly desemantized. It can be presumed that the value of ge- and
these other preverbs corresponds with that of ModHG preverbs such as be- or er-.
6. The reason why ge- and its cognates in the other West and East Germanic languages at-
tained such a prominent position in derivational morphology is unclear. Presumably there
was a choice of more or less desemantized preverbs, of which the most meaningless was

47
Quoted after LINDEMANN 1970: 20.
Page 22 of 27
Aspect in Old English Marc Zumstein

spread most easily in these derivational patterns. In Old Norse, on the contrary, the most
eminent preverb is of-, whereas the cognate of ge- plays only a marginal role.

Page 23 of 27
Aspect in Old English Marc Zumstein

10. REFERENCES

Barber, Charles (1993) The English language - a historical introduction Cam-

bridge: Cambridge University Press
Bloomfield, Leonard (1929) “Notes on the Preverb ge- in Alfredian English”, in:
Studies in English Philology: A Miscellany in Honor of
Frederick Klaeber (ed. Kemp Malone and Martin B.
Ruud), pp. 79 – 102, Minneapolis: The University of
Minneapolis Press
Bulygina, T. V. (1982) “K postroeniju tipologii predikatov v russkom jazyke”
(Towards a typology of predicates in the Russian lan-
guage) in: Seliverstova O. N. (ed.): Semantičeskie tipy
predikatov (Semantic types of predicates), pp. 7 – 85,
Moskva: Nauka
Comrie, Bernard (1976) Aspect – An Introduction to the Study of Verbal Aspect
and Related Problems, Cambridge: Cambridge Univer-
sity Press
Dürmüller, Urs and Utz, Hans (1977) Altenglisch – Eine Einführung, Tübingen: Max Niemeyer
Verlag
Finegan, Edward (1994) Language: Its Structure and Use, Cambridge: Cam-
bridge University Press
Guiraud-Weber, Marguerite (1993) “Le préverbe po- en russe moderne” in: La Revue Russe
5, pp. 57 – 68, Paris
Isačenko, A. V. (1962) Die russische Sprache der Gegenwart, Teil 1 – Formen-
lehre, Halle (Saale): Max Niemeyer Verlag)
Kluge, Friedrich (1975) Etymologisches Wörterbuch der Deutschen Sprache
(21st unchanged edition), Berlin, New York: Walter de
Gruyter
Lawson, Richard H. (1965) “The Prefix gi- as a Perfectivizing Future Significant in
OHG”, in: Journal of English and Germanic Philology 64,
pp. 90 – 97, Urbana: The University of Illinois Press
Lindemann, J. W. Richard (1969) “Old English Preverbal ge-: a Re-Examination of Some
Current Doctrines”, in: Lass, Roger (ed.): Approaches to
English Historical Linguistics - An Anthology, pp. 258 –
275, New York: Holt, Rinehart and Winston
Lindemann, J. W. Richard (1970) Old English Preverbal Ge-: Its Meaning, Charlotteville –
The University Press of Virginia
Mitchell, Bruce (1985) Old English Syntax (Volume I), Oxford: Clarendon Press
Mitchell, Bruce and Robinson, Fred (1992) A Guide to Old English (5th edition), Oxford: Blackwell
Publishers
Pilch, Herbert (1953) “Das AE. Präverb ge-”, in: Anglia 71, pp. 129 – 139, Tü-
bingen: Max Niemeyer Verlag

Page 24 of 27
Aspect in Old English Marc Zumstein

Pilch, Herbert (1955) “Der Untergang des Präverbs ge- im Englischen”, in:
Anglia 73, pp. 37 – 64, Tübingen: Max Niemeyer Verlag
Pilch, Herbert (1970) Altenglische Grammatik, München: Max Hueber Verlag
Pinsker, Hans (1976) Altenglisches Studienbuch (unter Mitwirkung von Udo
Fries und Peter Bierbaumer), Bern, München: Francke
Verlag
Quirk, R., Greenbaum, S. (1985) A comprehensive grammar of the English language,
Leech, G. and Svartvik, J., London: Longman
Raith, Josef (1951) Untersuchungen zum englischen Aspekt – I. Teil:
Grundsätzliches Altenglisch, München: Max Hueber
Verlag
Stanley, E. G (1973) “Discussion of: Lindemann, J. W. Richard (1970): Old
English Preverbal Ge-: Its Meaning” in: Anglia 91, pp.
493 – 494, Tübingen: Max Niemeyer Verlag
Streitberg, Wilhelm (1920) Gotisches Elementarbuch (5. and 6. edition), Heidel-
berg: Carl Winter’s Universitätsverlag, (Chapter 22, “III.
Aktionsarten”, pp. 194 – 200)
Vendler, Zeno (1967) Linguistics in philosophy, New York: Cornell University
Press (Chapter 4, “Verbs and Times”, pp. 97 – 121)
Weddige, Hilkert (1998) Mittelhochdeutsch – Eine Einführung, München: Verlag
C. H. Beck
Weimann, Klaus (1995) Einführung ins Altenglische, Heidelberg, Wiesbaden:
Quelle & Meyer

Page 25 of 27