This version of Favourite Articles has been archived and won't be updated before it is permanently deleted.
Please consult the revamped version of Favourite Articles for the most up-to-date content, and don't forget to update your bookmarks!
"Must" is an auxiliary verb which can express: a) a necessity; b) an insistent demand or a firm resolve; c) the inferred or presumed certainty of a fact; d) prohibition (when used negatively).
a) "Must" expresses necessity and is usually equivalent to "am (is, are) obliged". In the second person, it is used chiefly to express a command, or an insistent request or counsel:
In the third person, it tends to be restricted to the expression of a necessity which is either imposed by the will of the speaker, or relative to some specified end, or enunciated as a general proposition:
That necessity can take the form of a fixed or certain futurity. In this case "must" is equivalent to "fated or certain to":
Although less emphasized, the idea of necessity is also present in expressions such as "I must say", which means "I cannot help saying", and "you must know or understand" which means "you ought to be informed" or "I find it necessary that you should know or understand":
b) In the first person, "must" often expresses an insistent demand or a firm resolve on the part of the speaker:
It is also used in the second and third persons to render sentiments imputed to others:
c) Very often "must" expresses the inferred or presumed certainty of a fact, in other words, what Zandvoort calls "an assumption or conclusion",1 and what Poutsma calls "Predication of conviction".2 "Must" is the auxiliary by which the speaker expresses that "his mental attitude toward the fulfilment of the action or state expressed by the predicate… is one of conviction arrived at by the process of reasoning".3 It refers either to the present, and then it is followed by a present infinitive, as in "you must be aware of this", i.e. "I cannot doubt that you are aware of this"; or to the past, and then it is followed by a perfect infinitive, as in "he must have done it", i.e. "It is to be concluded that he did it."
Sometimes "must" is used to express an inference which will be rendered necessary if some particular assumption is made:
d) When used negatively, "must" expresses prohibition. According to the OED, "I must not" can mean "I am not allowed to", "I am obliged not to", etc., or "I will not permit myself to":
The adjunction of the negative brings about a considerable shift in the meaning of "must", so that "must not" is not the opposite of "must", but the opposite of "may". The absence of necessity is expressed by "need not". So the opposite of "you may come early tomorrow" will be "you mustn’t come…," whereas the opposite of "you must come early tomorrow" will be "you need not come early tomorrow".
"Must" is used as a present tense, and, under certain conditions, as a preterite. It is one of a separate group of anomalous verbs which is formed, according to Poutsma, "by the so-called preterite-present verbs, characterized by not having a personal ending in the third person singular of what is now used as a present, viz. can, dare, may, must, shall, will. Except for "dare", they are all of them defective in not having an imperative, nor any of the verbals".4 It can also be regarded as one of a small group of verbs, including "need" and "dare", "which have in various ways become partly or completely identical in the two tenses, present and preterite, and which have also to some extent the function as ’auxiliaries’ in common".5
Grammarians generally agree that "must" is an old past subjunctive which has gradually been losing its power to point to the past. "It is through the preterite of imagination", says Jespersen, "that ’must’ has become a present tense". The OED seems to agree with Jespersen when it writes: "The use of a present arose from the practice of employing the past subjunctive as a moderate, cautious, or polite substitute for the present indicative."
The use of "must" as a present (or future) does not seem to have caused any concern among grammarians. The same can be said of some of its uses as a preterite. Most grammarians seem to agree with the OED that, in modern use, "must" as a past tense is "confined to instances of oblique narration, and of the virtual oblique narration in which the speaker has in his mind what might have been said or thought at the time." In other words, when the clause containing ’must’ is subordinate either actually or in thought:
Jespersen states that "this is practically the only way in which ’must’ can be used as a preterite in modern colloquial speech."6
Beside this preterite which Jespersen calls "preterite of indirect speech", there is the "preterite of reality". This is where grammarians do not agree with each other. The OED states that there is one case when "must" can be used as a preterite without being part of a subordinate clause: "As a past or historical present tense, ’must’ is sometimes used satirically or indignantly with reference to some foolish or annoying action or some untoward event."
But, according to Jespersen and many others before him and after, the sentence of the type just mentioned are not the only ones where "must" is used as a preterite. Jespersen expressly states that "though this dictum of the great authority is, of course, substantially true, some examples of ’must’ to denote a real past time have been collected by Stoffel".7 C. Schulze went even farther and stated that the use of "must" as a preterite of reality is "far from being rare or obsolete", that "it occurs much more often than ’was (were) obliged, forced, etc." and is deliberately used to express all shades of the notion of necessity.8
Another grammarian, Klapperich, pronounced Schulze’s conclusions to be erroneous, saying that the majority of his examples were "dependent declarative clauses" and that "many of the seemingly independent clauses" contained "a latent oratio obliqua". He suggested that the statement that "must" as a preterite is rare "needs the addition of the qualification: in an independent sentence".9 But another grammarian, Klinghardt, saw no reason why "must" should be considered rare even in independent sentences. He firmly stated that he would not find anything strange in such sentences as: "Last night he must again go to bed rather early" thus contradicting Dr. Bradley who had laid down as a principle that to say ’I must go to London yesterday’ would now be "a ludicrous blunder".
In a recent study of this problem, Satoshi Ono, a Japanese linguist, sums up his findings by saying, first, "that in the case where ’must’ means inferred certainty or logical necessity, followed by a perfect infinitive, ’must’ can be regularly used in principal clauses of a conditional complex".10 In support of this he brings out one of the examples cited by Schulze:
and two cited by Gerbert:
This is perfectly in line with what is stated in the OED which gives these examples of ’must’ as a past conditional followed by a perfect infinitive:
Mr. Ono’s second conclusion is that "must" is "regularly used in that-clauses",11 and that in some cases "it is even used in adjective and adverbial clauses, including elliptical ones":
His third conclusion is that" ’must’ is likely to be used as a preterite in a principal clause, when it is followed or preceeded by ’needs’, ’necessarily’, or ’of necessity’ probably influenced by such words. This form can be found in the Bible and many modern writers". And he adds: "This ’must’, as is the case with other usages of this word, is not used objectively or dispassionately, but subjectively and so emphatically, probably because it originates from the past subjunctive".12 Among the examples that he gives are the following:
Mr. Ono’s fourth conclusion is that "must", apart from the cases where it is used in oblique narration or as a historical present tense, does occur, though not very often. His examples are the following:
Most of these examples are the same as those found in Jespersen’s A Modern English Grammar, Part IV, p. 7.
To summarize all this, we could say that "must" as a preterite occurs, not in completely independent sentences, but in sentences with given situations, i.e. usually in clauses subordinate not only in form but in sense. The problem then remains to determine whether sentences including "must" are completely independent sentences, or sentences where the verb is subordinate in thought to a verb in past time, or sentences in the virtual oblique narration, showing not a fact, but the speaker’s thought or soliloquy. In many cases it is almost impossible to determine in which sense the writer or speaker has used the expression. This may explain to a certain extent the differences of opinion among grammarians as to this or that particular example. So one can easily imagine that the problem of the preterite use of ’must’ will be something of an inexhaustible source of discussion, unless some grammarians find an adequate theory of the modal auxiliary verbs which will provide a means of settling a problem that seems insoluble on the level of discourse.
"Need" according to all dictionaries is a verb that means "have need of, want, require" and is conjugated as all full verbs normally are: "He needs money; do you need anything else?" But apart from this, it also has the meaning of "to be under a necessity or obligation to do something". When used in this sense, it often shows some peculiarities by which verbs are recognized as auxiliaries in English:
These facts should suffice to place "need" unmistakably among the auxiliaries. However, "need" is peculiar in that it also has parallel non-auxiliary uses:
This raises the problem of a possible difference of meaning between the two forms. Here is how Dwight L. Bolinger, in an article published in Vol. 4 of College English, approaches the problem:
Comparing the two (uses) we find another condition for auxiliary use satisfied—that expressed by the Merriam-Webster as "usually with some loss of its own original signification". The non-auxiliary "He needs to see the doctor" suggests a requirement close to the literal meaning of "need"; "Need we see the doctor?" suggests a more indefinite compulsion. Make the two phrases in all other respects identical, and the contrast is clear. "Does he need to see the doctor now?" implies a need for medical attention; "need he see the doctor now?" may imply "if he is to see him at all", or a number of other notions, but always something different from the inflected, non-auxiliary "does need", and tending toward the idea of external constraint rather than that of literal requirement or expression of lack within one’s self. The non-auxiliary calls for the fulfilment of a personal need, the auxiliary for compliance with something external.13
The Thorndike-Barnhart Comprehensive Desk Dictionary solves the problem in a completely different manner. According to this dictionary the difference between the two uses would simply be one of level of formality: "Note that ’need’ with the meaning of "must", "should", "have to" in rather formal usage is used as an auxiliary—that is, like "must" and "should" it has the third person singular form "need" (rather than ’needs’), it takes an infinitive without "to", and in negative and interrogative constructions is used as the main verb, not as an infinitive with ’do, does’. Examples of typical use are:
According to Bolinger, "need" is used in the following types of sentences, "the third of which", he states, "Curme misses".14
This seems to be in accordance with the OED, where it is considered that "need" is an auxiliary "when the clause has the forms ’it (she, I, etc.) Need not’, ’(why) need (it, etc.)’ or is virtually equivalent to one of these". What Bolinger achieves with his third category merely comes down to determining what can be considered "virtually equivalent" to the negative use of "need". It is not likely that Curme, who, according to Bolinger, "has dealt rather fully with the question", misses anything here. He probably considered that minimizing adverbs could be regarded as, in a way, equivalent to negations. This is exactly what Poutsma does when he includes negative contexts "such as imply a negative, although containing no negative word".15
Although all the examples given above are in the present indicative, "need" can also be used with all the characteristics of an auxiliary in the preterite indicative and the preterite conditional. And then, just like "must", it will have the same form for the preterite as for the present.
This is especially the rule in the preterite indicative in subordinate clauses, especially subordinate statements, in negative contexts:
In many cases the principal sentence is merely understood:
Exceptions, however, are not unfrequent:
"Need" is also used in principal sentences, but according to Poutsma, "is usual only when another verb shows the time sphere, but ’needed not’ is in ordinary language replaced by ’did not need’".16
"Need" is also practically used regularly, in negative context, in the preterite conditional followed by a perfect infinitive:
"Dare" is another verb that can be used with all the characteristics of an auxiliary, when it means "to have boldness or courage (to do something)", "to be so bold as". It is often followed by an infinitive without "to":
It can contract with the negative:
It can have a defective inflexion:
It does not always require "to do" in interrogative sentences:
There seems to be a large variety in the usage of "dare". This is made obvious in the OED where we read: "The original third person singular ’he dare’ and past tense ’durst’, remained undisturbed to the modern period, in which the transitive senses were developed; but early in the 16th Century, the new forms ’dares’ and ’dared’ appeared in the south, and are always used in the transitive sense, and now also in the intransitive sense when followed by ’to’. In the original construction, followed by the infinitive without ’to’, ’dare’ and ’durst’ are still in common use (especially in the negative ’he dare not’, ’he durst not’); and most writers prefer ’he dare go’ or ’he dares to go’ to ’he dares go’. The northern dialects generally retain ’he dare, he durst’, and writers of northern extraction favour their retention in literary English when followed by the simple infinitive without ’to’".
Jespersen does not agree with the OED as to the use of "durst" which, according to him, has become obsolete".17 Zandvoort also seems to consider this form obsolete since he does not even mention it in his Handbook of English Grammar. However he agrees that usage varies considerably in the use of "dare". Few of the rules he mentions in his book are presented as the only correct usage. He expressly states that "as to the use or absence of "to do", with "to dare" there are three possible constructions in negative sentences with ’not’: He dare(d) not return; He does (did) not dare to return; he does (did) not dare return".18 Speaking of the infinitive after "dare" when "dare" is preceded by an auxiliary, the most he can state is that "the infinitive is more commonly used with "to" than without".19 The only strict rules that can be found regarding "dare" are the one where Zandvoort expressly states that "after the present and the past tense of ’to dare’, the plain infinitive is always used in interrogative sentences and in negative sentences with enclitic ’not’ that are formed without the auxiliary ’to do’20 and the one where he says that after "daring" the infinitive is always used with "to".21
("Daring", though, cannot be used in the progressive form because it is an auxiliary like "must", "can". Zandvoort probably refers here to the "full" verb "to dare"!)
Another thing which must be mentioned in connection with "dare" is what the OED calls the "careless use" of the present "dare" for the past "dared". This anomalous use of "dare" is mentioned by Zandvoort and Jespersen and appears to be confined to the negative. "’Dare not’", says Zandvoort, "is usually pronounced [de[de:ant], and sometimes written without the second ’d’, when the time-sphere is sufficiently clear from the context".22 He gives the following examples:
Jespersen adds examples without a following "not" which, he says, "are less frequent, but nowadays not at all rare".23
To this Jespersen adds a mention of the imaginative use of "daren’t" as equivalent to "daredn’t":
CURME, George O., Parts of Speech and Accidence, D.C. Health and Company, Boston, 1935.
JESPERSEN, Otto, A Modern English Grammar, Part IV, GeorgeAllen and Unwin Ltd, London, 1949.
POUTSMA, H., A Grammar of Late Modern English, Part II: The Parts of Speech, Ed. p. Noorhoff, Groningen, 1926.
ZANDVOORT, R.W., A Handbook of English Grammar, Longmans, 1962.
Studies in English Grammar and Linguistics, A Miscellany in Honour of Takanobu Otsuka, Tokyo, 1958.
American Speech, Vol. 30, 1955. Crowell, F.L., ’Predating "have to", "must"?’ pp. 68-69.
College English (Chicago-Champaign) Vol. 4, 1942-43, Bolinger D.L., "Need: auxiliary", pp. 62-5.
© Public Services and Procurement Canada, 2024
TERMIUM Plus®, the Government of Canada's terminology and linguistic data bank
Writing tools – Favourite Articles
A product of the Translation Bureau