Tenses of English

What is a Tense?

As pointed out by Klein (1995: 141), 2500 years of research have not led to any precise or universally acknowledged definition of the category 'tense'. However, there is a widespread understanding of what 'tense' basically means. Such an 'understood' definition is for instance given in the International Encyclopedia of Linguistics (Bybee 1992: 223f.):

Tense refers to the grammatical expression of the time of the situation described in the proposition, relative to some other time. This other time may be the moment of speech, e. g. the past and future designate time before and after the moment of speech, respectively [...]. Tense is expressed by inflections, by particles, or by auxiliaries in construction with the verb [...].

Different Approaches

How many tenses are there in English? This is a simple question, to which, however, almost every linguist gives a different answer. According to Bublitz (1995: 135), these “dissenting views are reflected by the fact that linguists have yet to agree on the number of tenses [...] in English, which vary according to whether formal or semantic criteria or a combination of both are recognized. In particular, the present perfect, the past perfect and the future perfect defy easy categorization.” Linguists' different views on the number of English tenses can be grouped into four different approaches (cf. König 1995).

Two Tense Approach

Linguists who favor a binary approach argue that different tenses have to be distinguished morphologically, that is they have to be reflected in inflectional morphology. Accordingly, there are only two tenses, namely ‘past’ and ‘non-past’, as this is the only distinction that is reflected morphologically.

(1) He work - s.

(2) He work - ed.

All other categories of the English tense system are constituted by a combination of one of the two above-mentioned tenses with either a modal verb, or the perfect marker have. The advocates of this approach often conveniently stress the “combinatorial character” (König 1995: 154) of these members by naming them

• present perfect
• past perfect
• future perfect

rather than

• perfect
• pluperfect
• future II,

which might rather be used by linguists who recognize these as 'basic' tenses,. The binary approach is found in “many structuralist accounts” (König 1995: 154) such as the comprehensive grammar by Quirk et al. (1985).

Three Tense Approach

Some linguists have argued against regarding a sentence like We will/'ll miss the train as a combination of the present + the verb will + the infinitive miss. König (1995: 154) considers such an interpretation “untenable and synchronically inadequate” , as will has lost both morphological substance (it may be reduced to 'll) and semantic substance (since it no longer solely expresses volition, is no longer restricted to human beings and combines freely with the progressive and have). These developments allow for the interpretation of will as a future tense marker. Will has undergone a process of grammaticalization, during which a new grammatical category has developed. This view of the existence of three tenses (present, past, future) can be found in Klein (1994) and Hatav (1993), among other authors.

Six Tense Approach

In those approaches where six English tenses are distinguished, the list of tenses in most cases includes the following (cf. König 1995: 155):

• present
• past
• future tense
• present perfect (= perfect)
• past perfect (= pluperfect)
• future perfect (= future II)

This model is often based on the model of Latin grammar and on the assumption that combinatorial tenses such as the last three are only apparently compositional, i.e. from a morphological point of view, while being largely non-compositional semantically (“the combinatorial formal make-up of present perfect, past perfect, future perfect has no parallel in a compositional derivation of their meaning”, König 1995: 155).

The more-than-two tense approach is reflected in simple terms such as perfect and pluperfect (cf. Comrie 1985).

Eight Tense Approach

In addition to the six tenses mentioned above, Declerck (1990) includes two more tenses: the conditional tense and the conditional perfect. Declerck argues against a compositional analysis of combinatorial tenses. Aiming to account for the use of tenses in narrative discourse, he assumes two further tenses which, in his view, cannot be subsumed under any other category:

(3) At 5 o'clock, he would have left the hotel.

Historical Development

The evolution of a future and perfect tense (as acknowledged by some linguists) represents the most significant innovation of Modern English in comparison to earlier stages of that language. Old English had a two tense system (past and non-past; cf. Section 2.1), which is claimed to have been preserved in Modern English by some linguists. Crosslinguistic work (Bybee/Dahl 1989) has shown that future tense markers typically derive from:

• verbs of volition
• verbs of obligation
• verbs of motion

During a process of grammaticalization such verbs lose specific aspects of their meaning, and thus also broadening their distribution (e. g. to human beings) and/or morphological properties:

(4) It is going to rain.

(5) It's gonna rain.

Words may, in the process of grammaticalization, also acquire semantic substance. In the development of future tense markers it is the element of prediction and intention that is newly acquired. In specific contexts (e. g. future time reference), the choice of lexical or grammatical future tense marking is obligatory (e. g. will and going to), lending support to the view that the grammaticalization in this area is advanced. The expression that has advanced furthest as a future tense marker in English is will, as it combines freely with any type of verb and aspect.

The perfect is another example of the latest developments of the English tense system. Comparative linguists have identified typical historical sources of perfect markers, such as the verb have and the adverb already (König 1995: 164). The English perfect has developed from originally resultative sentences with be (John is gone) and have (I have the enemy bound). Such resultative sentences express states and/or possession over states. The change from resultative meaning to perfect meaning comes about due to a semantic change “as a result of which the responsibility for the action leading to the state is ascribed to the subject” (König 1995: 164). In other languages (though not in English), the perfect is also used as a narrative tense, thus gradually replacing (and presumably eventually eliminating) the past tense (preterite).

Declerck's Theory of Tense

This section provides a brief description of the temporal system proposed by Declerck (1991, 2006). For the sake of comprehensibility, Declerck's terminology and basic assumptions will be briefly outlined.

Terminology

Declerck (2006: 22) conceives of tense as a “linguistic concept [which] denotes the form taken by the verb to [...] express the temporal relation between the time of the situation in question and an 'orientation time' which may be either the 'temporal zero-point' (which is usually the time of speech [...]) or another orientation time that is temporally related to the temporal zero-point.” The orientation time is “[a]ny time that can provide the 'known' time (or one of the known times) required for the expression of the temporal relation(s) encoded in a tense form” (Declerck 2006: 117). The temporal zero-point (t0) is the point in time from which all expressed temporal relations take their starting point. It is usually (but not necessarily) the time of the utterance.

In (6) below, the orientation time corresponds to t0, whereas in (7), the past tense form confessed locates the time of the confession in the past, and the past perfect form had stolen expresses that the situation (namely the theft) was committed even before the confession. In the latter case, the time of the confession is thus the orientation time for the past perfect form, which in turn lies before (and therefore does not equal) t0. The form had stolen locates the situation relative to an orientation time which is itself located relative to speech time (=t0).

(6) I met him last night.

(7) Last night he confessed that he had stolen the money.

Temporal Relations – Absolute and Relative Tenses

Declerck differentiates between absolute and relative tenses. Absolute tenses are those which express “a direct temporal relation with the temporal zero-point (= t0)” (Declerck 2006: 25), just as in (6). Relative tenses are those which express “a single temporal relation between the time of the situation referred to and an orientation time other than the zero-time (= t0)” (ibid.), just like in (7). Relative tenses express one of the following temporal relations:

• anteriority: the situation time precedes the orientation time: He said he had got up early.
• simultaneity: the situation time coincides with the orientation time: He said he didn't feel well.
• posteriority: the situation time follows the orientation time: He said he would save us.

Absolute tenses are: present, past, present perfect, future.

Relative tenses are: past perfect, conditional.

The future perfect and the conditional perfect express two temporal relations at the same time: “the time of the situation is represented as anterior to an orientation time which is itself represented as posterior to another time” (Declerck 2006: 25). In the case of the future perfect, this 'other time' is t0, which makes the future perfect an absolute-relative tense, because it relates its situation time to an orientation time (this is the relative component) which is itself related to t0 (this is the absolute component). With the conditional perfect, neither of the orientation times is t0, which means that this tense has two relative components; Declerck calls this tense a 'complex relative tense' (2006: 25).

Tenses

The only two tenses which can be differentiated morphologically are the present and the past. All other tenses are formed with the help of auxiliaries. These other tenses are called 'complex tenses', where the first auxiliary (= the operator, i.e. either have or will) is morphologically in the present or past tense (cf. Declerck 2006: 24).

All tenses which show present tense inflection (present perfect, future, future perfect) create a present time-sphere,. All tenses exhibiting past tense morphology (past perfect, conditional tense, conditional perfect) create a past time-sphere (cf. Declerck 2006: 156). Declerck specifies the concept of ‘time-sphere’ as follows: English tenses “reflect a mental division of time into past and nonpast. The main evidence for this is that all tenses carry either a past or a nonpast (present) tense morpheme. There is no future tense morpheme” (2006: 147). Declerck represents this mental division of time as two time-spheres (cf. above), in which the different tenses are situated.

The present time-sphere contains the present zone (which coincides with t0), the pre-present zone (which leads up to t0), and the post-present zone (which begins immediately after t0). In order to locate a situation in one of these zones, the present, present perfect or future are used.

The past time-sphere lies wholly before t0. In order to locate a situation in the past time-sphere, the past perfect, conditional tense or conditional perfect are used. The past time-sphere is conceived of as disconnected from the present time-sphere.

Present

The forms of the present of all verbs (except be and have) are homophonous with the stem of the verb (e. g. forget), except in the third person singular (e. g. forgets). The meaning of the present tense is that “the situation time (i. e. the time of the predicted situation) coincides with t0” (Declerck 2006: 173). Since t0 is a point in time, the situation time is also a point in time. The full situation expressed by the tense form takes one of the following forms: (i) One possibility is that the entire situation is punctual. In this case the situation time coincides with the time of the entire situation and is thus punctual, too.

(8) [Now watch closely.] I hit the nail once with a hammer.

(ii) Another possibility is that the entire situation is durative. In this case, the situation time cannot be located at (punctual) t0. Thus, if the entire situation is durative, the situation time “can be a punctual subinterval of the time of the full situation and that punctual subinterval can be located at t0” (Declerck 2006: 173). This is only possible if the entire situation is homogeneous, which means that the sentence denoting the situation can be used not only to describe the situation as a whole, but also each portion of it (e. g. in the durative process of working, each 'subpart' of working is also working).

(9) I am working.

Sentence (9) is homogeneous: if I am working for a period of time, then this sentence is a valid description of any part of the situation that coincides with t0.