BASIC QUALITIES OF THE PERFECT FORMS
The Modern English perfect forms have been the subject of a lengthy discussion which has not so far brought about a definite result. The difficulties inherent in these forms are plain enough and may best be illustrated by the present perfect. This form contains the present of the verb have
and is called present perfect, yet it denotes an action which no longer takes place, and it is (almost always) translated into Russian by the past tense, e. g. has writ
ten
— написал, has arrived
— приехал, etc.
The position of the perfect forms in the system of the English verb is a problem which has been treated in many different ways and has occasioned much controversy. Among the various views on the essence of the perfect forms in Modern English the following three main trends should be mentioned:
1. The category of perfect is a peculiar tense category, i. e. a category which should be classed in the same list as the categories "present" and "past". This view was held, for example, by O. Jespersen. 1
2. The category of perfect is a peculiar aspect category, i. e. one which should be given a place in the list comprising "common aspect" and "continuous aspect". This view was held by a number of scholars, including Prof. G. Vorontsova.2
Those who hold this view have expressed different opinions about the particular aspect constituting the essence of the perfect forms. It has been variously defined as "retrospective", "resultative", "successive", etc.3
3. The category of perfect is neither one of tense, nor one of aspect but a specific category different from both. It should accordingly be designated by a special term and its relations to the categories of aspect and tense should be investigated. This view was expressed by Prof. A. Smirnitsky. He took the perfect to be a means of expressing the category of "time relation" (временнаяотнесенность).4
This wide divergence of views on the very essence of a verbal category may seem astonishing. However, its causes appear to be clear enough from the point of view of present-day linguistics. These causes fall under the following three main heads:
1
See O. Jespersen, The Philosophy of Grammar,
p. 254 ff.
2
See Г. Н. Воронцова, Очерки по грамматике английского языка,
1960. стр. 191 сл.
3
Ibid.
4
See А. И. Смирницкий. Перфект и категория временной отнесённости.
Иностранныеязыкившколе, 1955, №
1, 2.
Basic Qualities of the Perfect Forms
91
1. Scholars have been trying to define the basic character of this category without paying sufficient attention to the system of categories of which it is bound to make a part. As we shall see presently, considerations of the system as a whole rule out some of the proposed solutions.
2. In seeking the meaning of the category, scholars have not always been careful to distinguish between its basic meaning (the invariable) and its modifications due to influence of context.
3. In seeking the basic meaning of the category, scholars have not always drawn a clear line of distinction between the meaning of the grammatical category as such and the meanings which belong to, or are influenced by, the lexical meaning of the verb (or verbs) used in one of the perfect forms.
If we carefully eliminate these three sources of error and confusion we shall have a much better chance of arriving at a true and objective solution. Let us now consider the views expressed by different scholars in the order in which we mentioned them above.
If we are to find out whether the perfect can be a tense category, i. e. a tense among other tenses, we must consider its relations to the tenses already established and not liable to doubts about their basic character, i. e. past, present, and future. There is no real difficulty here. We need only recollect that there are in Modern English the forms 1
present perfect, past perfect, and future perfect. That present, past, and future are tense categories, is firmly established and has never been doubted by anyone. Now, if the perfect were also a tense category, the present perfect would be a union of two different tenses (the present and the perfect), the past perfect would likewise be a union of two different tenses (the past and the perfect) and the future perfect, too, would be a union of two different tenses (the future and the perfect). This is clearly impossible. If a form already belongs to a tense category (say, the present) it cannot simultaneously belong to another tense category, since two tense categories in one form would, as it were, collide and destroy each other. Hence it follows that the category of perfect cannot be a tense category. We need not consider here various views expressed by those who thought that the perfect was a tense, since their views, whatever the details may be, are shown to be untenable by the above consideration. So the view that the perfect is a special tense category has been disproved.
In order to find out whether the perfect can be an aspect category, we must consider its relations to the aspects already established,
1
We use here the non-committal term "form" to avoid any pre-judgement concerning the essence of the category in question. We will use the term in similar contexts elsewhere.
92The Verb: The Perfect
viz. the common and the continuous aspects. 1
This problem does not present any particular difficulty, either. We need only recollect that there are in Modern English such pairs as is writing
— has
been writing, was writing
— had been writing, will be writing
— will have been writing,
i. e. present continuous and present perfect continuous, past continuous and past perfect continuous, future continuous and future perfect continuous. All of these forms belong to the continuous aspect, so the difference between them cannot possibly be based on any aspect category. For example, since both was writing
and had been writing
belong to the continuous aspect (as distinct from wrote
and had written),
they cannot be said to differ from each other on an aspect line; otherwise they would at the same time belong to one aspect and to different aspects, which is obviously impossible. Hence the conclusion is unavoidable that the perfect is not an aspect. The views of those who consider the perfect to be an aspect need not therefore be discussed here in detail.
Since the perfect is neither a tense nor an aspect, it is bound to be some special grammatical category, different both from tense and from aspect. This view, though not quite explicitly stated, was first put forward by Prof. A. Smirnitsky in a posthumous article.2
It is in complete harmony with the principle of distributive analysis, though Prof. Smirnitsky did not, at the time, use the term "distributive analysis".
The essence of the grammatical category expressed by the perfect, and differing both from tense and from aspect, is hard to define and to find a name for. Prof. Smirnitsky proposed to call it "the category of time relation", which is not a very happy term, because it seems to bring us back to the old view that the perfect is a special kind of tense — a view which Prof. Smirnitsky quite rightly combated. Later it was proposed to replace his term of "time relation" by that of "correlation" (соотнесенность), which has the advantage of eliminating the undesirable term "time". This is decidedly the term to be preferred.
As to the opposition in such pairs as writes
— has written,
wrote
— had written, will write
— will have written, is writing
— has been writing, was writing
—
had been writing, will be writ
ing
—
will have been writing,
Prof. Smirnitsky proposed to denote it by the correlative terms "non-perfect" and "perfect". While this
1
We are proceeding here on the assumption that the existence of these two aspects, and, indeed, of aspect as a category of the English verb has been recognised. If its existence is denied the problem presents itself in a different light (see p. 81).
2
See А. И. Смирницкий, Перфект и категория временной отнесён
ности.
Иностранные языки в школе, 1955, № 2. Seealso А. И. Смирницкий, Морфология английского языка,
1959, стр. 274—316. Compare И. II. Иванова, Вид и время в современном английском языке,
стр. 112—113.
Basic Qualities of the Perfect Forms
93
latter proposal may be fully accepted, the definition of the meaning of the category presents considerable difficulty. Its essence appears to be precedence: an action expressed by a perfect form precedes some moment in time. We cannot say that it always precedes another action: the present perfect form is most commonly used in sentences which contain no mention of any other action.
On the other hand, the use of a non-perfect form does not necessarily imply that the action did not precede some moment in time. It may, or it may not, have preceded it. To find this out, the reader or hearer has to take into account some other feature of the context, or, possibly, the situation, that is, an extralinguistic factor. Thus, the opposition between perfect and non-perfect forms is shown to be that between a marked and an unmarked item, the perfect forms being marked both in meaning (denoting precedence) and in morphological characteristics (have
+
second participle), and the non-perfect forms unmarked both in meaning (precedence not implied) and in morphological characteristics (purely negative characteristic: the collocation "have
+ second participle" not used). On the whole, as a general term to denote the basic meaning of the perfect the term "correlation" in the above-mentioned meaning seems quite acceptable and we propose to make use of it until a better term is found, which may take some time to happen.
If this view is taken, the system of verbal categories illustrated by the forms writes, is writing, has written, has been writing, wrote,
was writing, had written, had been writing, will write, will be writing, will have written, will have been writing,
—
is based on three groups of notions, viz. tense:
present vs. past vs. future; aspect:
common vs. continuous; correlation:
non-perfect vs. perfect. As is seen from this list, the latter two of the three oppositions are double (or "dichotomic"), i.e. they consist of only two items each, whereas the first (the tense opposition) is triple (or "trichotomic"), i. e. it consists of three items.
We will accept this state of things without entering into a discussion of the question whether every opposition must necessarily be dichotomic, i. e. consist of two members only.
Thus, the opposition between writes
and wrote
is one of tense, that between wrote
and was writing
one of aspect, and that between wrote
and had written
one of correlation. It is obvious that two oppositions may occur together; thus, between writes
and was
writing
there are simultaneously the oppositions of tense and aspect; between wrote
and will have written
there are simultaneously the oppositions of tense and correlation, and between wrote
and had been writing
there are simultaneously the oppositions of aspect and correlation. And, finally, all three oppositions may occur together: thus, between writes
and had been writing
there are simultaneously the oppositions of tense, aspect, and correlation.
94The Verb: The Perfect
If, in a system of forms, there is only one opposition, it can obviously be represented graphically on a line. If there are two oppositions, they can be represented on a plane. Now, if there are three oppositions, the system obviously cannot be represented on a plane.
However, a drawing of a parallelepiped cannot give the desired degree of clarity and we will not reproduce it here.
USES OF THE PERFECT FORMS
We have accepted the definition of the basic meaning of the perfect forms as that of "precedence". However, this definition can only be the starting point for a study of the various uses of the perfect forms. Indeed, for more than one case this definition of its meaning will seem wholly inadequate, because its actual meaning in a given context will be influenced by various factors. Though a very great amount of investigation has been carried on in this field and many phenomena have by now been elucidated, it is only fair to say that a complete solution of all the problems involved in the uses and shades of meaning of the perfect forms in Modern English is not yet in sight.
Let us first, ask the question: what kinds of linguistic factors can be expected to have an influence on the use and shades of meaning of the perfect forms? We will try to answer this question in a general way, before proceeding to investigate the possible concrete cases.
These factors, then, would seem to be the following:
(1) the lexical meaning of the verb;
(2) the tense category of the form, i. e. whether it is the present perfect, past perfect, or future perfect (we cannot be certain in advance that the tense relation is irrelevant here);
(3) the syntactical context, i. e. whether the perfect form is used in a simple sentence, or the main clause, or again in a subordinate clause of a complex sentence.
To these should be added an extralinguistic factor, viz.
(4)the situation in which the perfect form is used.
Let us now consider each of these factors separately and then come to the question of their possible interaction.
(1)The meaning of the verb used can affect the meaning of the perfect form in so far as the verb may denote either an action which is apt to produce an essential change in the state of the object (e. g. He has broken the cup)
or a process which can last indefinitely
1
See А. И. Смирницкий, Морфология английского языка,
стр. 310.
Uses of the Perfect Forms
95
without bringing about any change (e. g. He has lived in this city
since
1945),
etc. With the verb break,
for instance, the shade of meaning would then be the result of the action (the cup is no longer a cup but a collection of fragments), whereas with the verb live
no result in this exact sense can be found; we might infer a resultative meaning only in a somewhat roundabout way, by saying that he has now so many years of life in this city behind him. Thus the meaning of result, which we indeed do find in the sentence He
has broken the cup,
appears to be the effect of the combined meanings of the verb as such (in whatever form) and the perfect form as such. It is quite natural that this meaning should have more than once been taken to be the meaning of the perfect category as such, which was a misconception.1
To give another example, if the verb denotes an action which brings about some new state of things, its perfect form is liable to acquire a shade of meaning which will not be found with a verb denoting an action unable to bring about a new state. We may, for instance, compare the sentences We have found the book
(this implies that the book, which had been lost, is now once more in our possession) and We have searched the whole room for the book
(which does not imply any new state with reference to the book). Of course many more examples of this kind might be given. The basic requirement is clear enough: we must find the meaning of the form itself, or its invariable, and not the meaning of the form as modified or coloured by the lexical meaning of the verb. If this requirement is clearly kept in mind, many errors which have been committed in defining the meaning of the form will be avoided.
(2) The possible dependence of the meaning of perfect forms on the tense category (present, past or future) is one of the most difficult problems which the theory of the perfect has had to face. It is quite natural to suppose that there ought to be an invariable meaning of the phrase "have
+
second participle", no matter what the tense of the verb have
happens to be, and this indeed is the assumption we start from. However, it would be dangerous to consider this hypothesis as something ascertained, without undertaking an objective investigation of all the facts which may throw some light on the problem. We may, for instance, suspect that the present perfect, which denotes "precedence to the present", i. e. to the moment of speech, may prove different from the past perfect, denoting precedence to a moment in the past, or the future perfect, denoting precedence to a moment in the future: both the past and the future are, of course, themselves related in some way to the
1
This was very aptly pointed out by Prof. G. Vorontsova in her book (p. 196), where she criticised this conception of the English perfect found in several authors.
86The Verb: The Perfect
present, which appears as the centre to which all other moments of time are referred in some way or other. One of the chief points in this sphere is the following. If an action precedes another action, and the meaning of the verb is such a one that the action can have a distinct result, the present perfect form, together with the lexical meaning of the verb (and, we should add, possibly with some element of the context) may produce the meaning of a result to be seen at the very moment the sentence is uttered, so that the speaker can point at that result with his finger, as it were. Now with the past perfect and with the future perfect things are bound to be somewhat different. The past perfect (together with the factors mentioned above) would mean that the result was there at a certain moment in the past, so that the speaker could not possibly point at it with his finger. Still less could he do that if the action he spoke about was in the future, and the future perfect (again, together with all those factors) denoted a result that would be there in the future only (that is, it would only be an expected result). 1
All this has to be carefully gone into, if we are to achieve really objective conclusions and if we are to avoid unfounded generalisations and haphazard assertions which may be disproved by examining an example or two which did not happen to be at our disposal at the moment of writing.
(3) The syntactical context in which a perfect form is used is occasionally a factor of the highest importance in determining the ultimate meaning of the sentence. To illustrate this point, let us consider a few examples: There was a half-hearted attempt at a
maintenance of the properties, and then Wilbraham Hall rang with
the laughter of a joke which the next day had become the common precious property of the Five Towns.
(BENNETT) Overton waited
quietly till he had finished.
(LINDSAY) But before he had answered,
she made a grimace which Mark understood.
(R. WEST) The action denoted by the past perfect in these sentences is not thought of as preceding the action denoted by the past tense.
Another possibility of the context influencing the actual meaning of the sentence will be seen in the following examples. The question, How long have you been here?
of course implies that the person addressed still is in the place meant by the adverb here.
An answer like I
have been here for half an hour
would then practically mean, 'I have been here for half an hour and I still am here and may stay here for some time to come'. On the other hand, when, in G. B. Shaw's play, "Mrs Warren's Profession" (Act I), Vivie comes into the room and Mrs Warren asks her, "Where have you been,
Vivie?"
it is quite evident that Vivie no longer is in the place about
1
See also below (p. 111) on the modal shades of the future.
Uses of the Perfect Forms
97
which Mrs Warren is inquiring; now she is in the room with her mother and it would be pointless for Mrs Warren to ask any question about that. These two uses of the present perfect (and similar uses of the past perfect, too) have sometimes been classed under the headings "present (or past) perfect inclusive" and "present (or past) perfect exclusive". This terminology cannot be recommended, because it suggests the idea that there are two different meanings of the present (or past) perfect, which is surely wrong. The difference does not lie in the meanings of the perfect form, but depends on the situation in which the sentence is used. The same consideration applies to the present (or past) perfect continuous, which is also occasionally classified into present (or past) perfect continuous inclusive and present (or past) perfect continuous exclusive. The difference in the meaning of sentences is a very real one, as will
be seen from the following examples. "Sam, you know everybody," she said, "who is that terrible man I've been talking to? His name is Campofiore."
(R. WEST) I
have been saving money
these many months.
(THACKERAY, quoted by Poutsma) Do you mean to say that lack has been playing with me all the time? That he has been urging me not to marry you because he intends to marry you himself?
(SHAW) However, this is not a difference in the meaning of the verbal form itself, which is the same in all cases, but a difference depending on the situation or context. If we were to ascribe the two meanings to the form as such, we should be losing its grammatical invariable, which we are trying to determine.
Of course it cannot be said that the analysis here given exhausts all possible uses and applications of the perfect forms in Modern English. We should always bear in mind that extensions of uses are possible which may sometimes go beyond the strict limits of the system. Thus, we occasionally find the present perfect used in complex sentences both in the main and in the subordinate clause — a use which does not quite fit in with the definition of the meaning of the form. E. g. I've sometimes wondered if I haven't
seemed a little too frank and free with you, if you might not have thought I had "gone gay", considering our friendship was so far from intimate.
(R. WEST) We shall best understand this use if we substitute the past tense for the present perfect. The sentence then would run like this: I
have sometimes wondered if I hadn't seemed
a little too frank and free with you...
An important shade of meaning of the original sentence has been lost in this variant, viz. that of an experience summed up and ready at the time of speaking. With the past tense, the sentence merely deals with events of a past time unconnected with the present, whereas with the present perfect there is the additional meaning of all those past events being alive in the speaker's mind.
4 Б. A. Ильиш
98The Verb: The Perfect
Other examples might of course be found in which there is some peculiarity or other in the use of a perfect form. In the course of time, if such varied uses accumulate, they may indeed bring about a modification of the meaning of the form itself. This, however, lies beyond the scope of our present study.
The three verbal categories considered so far — aspect, tense, and correlation — belong together in the sense that the three express facets of the action closely connected, and could therefore even occasionally be confused and mistaken for each other. There is also some connection, though of a looser kind, between these three and some other verbal categories which we will now consider, notably that of mood and that of voice. We will in each case point out the connections as we come upon them.