Boogaart & Janssen’s (2007:808ff) review of non-temporal analyses of tense discusses a number of examples in the simple present in which the eventuality occurs before the time of utterance. Two of their examples, from English and Dutch respectively, are given as (378); the fact that the telling/asking “took place in the past” (i.e. before the time of utterance) is claimed to be a problem for temporal theories of tense, since “such discrepancies cannot be accounted for coherently in exclusively temporal terms”. From the point of view encoded in Figure 24: Simple tenses in Dutch, however, this claim is clearly premature, because simple present examples such as (378) are precisely what we expect to arise on our hypothesis that eventuality k can in principle be included anywhere in the present i of the speaker/hearer.

Simple present tense situations in which the eventuality k precedes or follows speech time n also arise when the speaker gives a second-hand report. If Els promised the speaker yesterday to read his paper today, the speaker can utter example (379) at noon to report this promise, even if Els has already read his paper in the morning or will start reading it later in the day.

That we can explain why the simple present can also refer to an imperfect eventuality preceding or following n on the assumption that Dutch does not express the binary feature [±posterior] within its verbal system provides strong support for the binary tense theory. This is especially true because Boogaart & Janssen are correct in claiming that it cannot be explained without additional stipulations by Reichenbachian approaches to the verbal tense system, or any other approach that equates the simple-present tense with speech time.

The default interpretation of example (379) can be overridden by pragmatic considerations. In the context given above, the split-off point of the possible worlds precedes present-tense interval i, and thus also precedes speech time n. If, however, the speaker is able to monitor Els’ actions, the split-off point of the possible worlds coincides with n, and in this case example (379) can only be used to refer to the situation depicted in Figure 26, in which eventuality k must follow or overlap with n.

The fact that the simple present cannot be used when the eventuality k precedes n is due entirely to pragmatics; since the speaker knows that k precedes n (i.e. that k is completed within the actualized part ia of the present-tense interval), he can describe this situation more precisely by using the present perfect (cf. Section 1.5.4.2, sub II/III), and Grice’s maxim of quantity therefore prohibits the use of the less informative simple present.

The interpretation of example (379) can also be restricted by grammatical means, such as the addition of temporal adverbial phrases, as in (380). Note in passing that, under the working assumption that the speech time is noon, (380a) is felicitous only if the split-off point of the possible worlds precedes speech time n; if the split-off point coincides with n, the present tense is excluded by the same pragmatic reasoning discussed in Subsection II.

The adverbial phrases vanmorgenthis morning, op dit momentat this moment and vanmiddagthis afternoon locate eventuality k before, simultaneously with or after n, i.e. noon; we illustrate this in Figure 27 for the adverbial phrase vanmiddagthis afternoon. The effect of the addition of a temporal adverbial phrase is thus that time interval j is restricted to a subpart of i, which is located either in the actualized part of the present/past-tense interval, in the non-actualized part of the present/past-tense interval, or in some other part of the present/past-tense interval that contains speech time n.

Temporal adverbial phrases do not necessarily restrict the present j of eventuality k, but can also modify k itself. This can be observed in example (381), where vanmiddagthis afternoon modifies j and na afloop van haar collegeafter her course modifies k; the event time interval k must be located within the time interval j denoted by vanmiddag and after the moment in time referred to by na afloop van haar college.

The effect of the adverbial modification of time interval k is particularly striking with momentaneous events such as bereikento reach in (382); this example asserts that in all possible worlds eventuality k is located within the interval j denoted by vanmiddagafternoon and includes 3 p.m. Since the eventuality is momentaneous, this implies that in all possible worlds the eventuality will occur at 3 p.m. (where the given time is intended and understood as an approximation: “at about 3 p.m.”).

It can be argued that the resulting reading of (382) is not due to the independent modification of the time intervals j and k, but that we are dealing with a single adverbial phrase vanmiddag om drie uur. That this is a possible analysis is undeniable, since the whole string can occur in clause-initial position, but example (383) shows that the proposed analysis is also a possible one: the string vanmiddag om drie uur can be split and the two parts are assigned different scopes with respect to the modal adverb waarschijnlijkprobably; cf. Section 8.2.3. for further discussion.

The adverb vanmiddag, the modifier of j, has wide scope with respect to the modal adverb; it is claimed that in all possible worlds the eventuality of reaching the finish will take place in the afternoon. The adverbial phrase om 3 uur, the modifier of eventuality k, on the other hand, has narrow scope with respect to the modal adverb; it is claimed that in the majority of possible worlds the eventuality of reaching the finish will take place at three o’clock. The result is that the speaker asserts that it is certain that the eventuality of the peloton reaching the finish will occur in the afternoon and that there is a high probability that the event time interval k contains the time denoted by the phrase om 3 uur. The fact that the string vanmiddag om drie uur can be split, and that the two parts can take scope independently of each other, is clear evidence that it does not have to form a single constituent, but can consist of two independent temporal adverbial phrases; cf. Section 8.2.3. for further discussion of the co-occurrence of multiple temporal adverbials in a single clause.

So far, we have tacitly assumed that the eventuality denoted by the lexical projection of the main verb occurs only once. Although this may be the default interpretation, the examples in (384) show that this is certainly not necessary: example (384a) expresses that the speaker will eat three times within present-tense interval i: once in the time interval j denoted by vanmorgenthis morning, once in the time interval j' denoted by vanmiddagthis afternoon, and once in the time interval j'' denoted by vanavondthis evening. Similarly, the frequency adverb vaakoften in (384b) expresses that within present-tense interval i (which in this case covers a longer period of months or even years) there are many instances of the eventuality denoted by the phrase naar de bioscoop gaango to the cinema.

The fact that the present/past-tense interval i can contain multiple occurrences of the eventuality denoted by the lexical projection of the main verb is exploited to the full in habitual constructions such as (385), in which a regularly occurring eventuality can be expressed without the use of an overt adverbial phrase. The availability of this reading may again be a matter of pragmatics, but there are also analyses that postulate empty operators with a similar function as frequency adverbs like altijdalways or vaakoften; cf. Oosterhof (2008) for examples of such proposals.

From habitual examples such as (385), it seems just a small step to get to truly generic examples such as (386); cf. Section N18.1.1.5 for a discussion of the different types of generic examples.

Note that examples similar to (385) and (386) can easily be given in the simple past. Even the past-tense counterpart of example (386c), i.e. De walvis was een zoogdierThe whale was a mammal, is possible with the reading that in a certain past-tense interval whales were mammals. The use of this sentence is of course infelicitous in our world, since it wrongly suggests that whales are not mammals in the present-tense interval (or that they are extinct), but this is again due to pragmatics: if the speaker is aware of the fact that whales are also mammals in the present-tense interval, Grice’s maxim of quantity would have required the use of the present tense with a present-tense interval that includes the past-tense interval.

Present-tense examples such as (387) allow at least two readings, which we will call conditional and hypothetical. This subsection shows that the choice between the two readings is pragmatic in nature.

The conditional reading is the default reading and expresses that for any subinterval of the present-tense interval for which it is true that the speaker has enough money, it will also be true that the speaker will go on vacation. The hypothetical reading is pragmatic in nature and arises when the actualized part of the present-tense interval is considered irrelevant: the utterance expresses that in any future world in which the speaker has enough money, he will go on vacation. The ambiguity between the two readings can be resolved by adverbial modification.

Modification of the consequence of the construction by a frequency adverb such as altijdalways favors the conditional reading, whereas modification of the antecedent by a temporal adverbial phrase such as volgend jaarnext year triggers the hypothetical reading. That this is more than just a tendency is shown by the examples in (389). Given that (389a) expresses an established fact of chemistry, it is only compatible with a conditional reading. This is reflected by the fact that it is easily possible to modify the consequence by a frequency adverb, but that modification of the antecedent by a time adverb leads to an infelicitous result.

Past-tense examples such as (390a) also allow at least two readings. The first is again conditional, but the second is counterfactual rather than hypothetical. We will argue below that the choice between the two readings is again pragmatic in nature. Note that examples such as (390b) are special in that the conditional reading is excluded: this is of course due to pragmatics, since it is a priori unlikely that the proposition in the antecedent of the construction (“I am you”) will be true in any possible world.

The conditional reading is again the default reading and expresses that for every subinterval in past-tense interval i of which it is true that the speaker has money, it is also true that the speaker is going on vacation. The counterfactual reading arises when the antecedent of the construction is not, or is not expected to be, fulfilled in the speaker’s actual world (within the relevant past-tense interval). First, if the condition expressed by the antecedent of the construction had been fulfilled in the speaker’s actual world before speech time n, the speaker could be more precise using example (391a). Second, if the speaker believes that the condition will be fulfilled in some possible future world, he can be more precise using example (391b).

The maxim of quantity thus leaves the addressee no other choice but to conclude that the speaker believes that the antecedent in (390a) is fulfilled only in possible worlds other than the actual one, which. moreover, must have a split-off point before n. This leads to the counterfactual interpretation.

An interesting fact about conditionals and counterfactuals is that als-phrases often alternate with constructions without als, in which the finite verb occupies the initial position of a main clause. Such verb-first constructions can be used to express wishes, especially when a particle such as maar is present; note that under the wish reading the consequence can easily be left implicit. This shows that the maxim of quantity is more generally applicable to derive irrealis constructions of various types.

Example (393a) shows that counterfactual interpretations also arise in examples with an epistemic modal verb in the past tense. This possibility follows from the assumptions made so far: the past tense on the modal verb in the first conjunct indicates that some source had reason to believe that the collapse of the house was inevitable, while the second conjunct indicates that this belief was wrong. Counterfactual readings are not possible in present-tense examples such as (393b); if the speaker and the addressee know that the house did not collapse before speech time n, a future interpretation emerges for the pragmatic reasons given in Section 1.5.2, sub I.

The past tense of the modal verb zullenwill is often used to express a counterfactual interpretation. Example (394a) is counterfactual for the same reason as (393a): the past tense of zullen indicates that to some source had information suggesting that the house would collapse, but the second conjunct again indicates that this assessment was wrong. The present-tense counterpart of this example in (394b) again has a future interpretation for pragmatic reasons; cf. Section 1.5.2, sub II, for further discussion.

Further discussion of the relationship between counterfactual interpretations and the past tense can be found in Section 1.4.3, sub II, where it is shown that many instances of the German past subjunctive can be expressed by regular past marking in Dutch.

Pragmatics can explain why the simple past can be used to express that a certain nominal description does not apply to a certain entity. Imagine the situation in which a pregnant woman gets on a bus. All the seats are taken and no one seems willing to oblige her by offering her a seat. An elderly lady gets angry and says (395) to the boy sitting next to her. Knowing that he had no intention of giving up his seat, she implies that the description een echte heer does not apply to him. This use of the simple past seems to be widespread in children’s language; examples such as (395b) introduce a play in which the participants assume certain model roles.

This section has shown that the default reading of the simple tenses is that the present j of eventuality k, i.e. the interval in which the eventuality denoted by the lexical projection of the main verb must take place, is identical to the complete present/past-tense interval: the eventuality can take place before, during, or after speech time n/n'. In many cases, however, the interpretation is more restricted and sometimes has non-temporal implications. We have shown in this section that this follows from the interaction between temporal information (e.g. expressed by tense and adverbial modification), modal information encoded in the sentence (i.e. the theory of possible worlds) and pragmatic information (i.e. Grice’s maxim of quantity).