Sources of Modal Necessity: The Case of Need to’

Abstract

This paper offers a principled framework for the analysis of sources associated with modal verbs that express root necessity. First, the notion of “source” (who or what lies at the origin of the necessity) is described and illustrated, and a comprehensive taxonomy of sources is put forward that can be used for empirical analysis. Special attention is paid to “subject-oriented” sources, which have mainly been discussed in the realm of modal possibility (ability) but which are relevant to modal necessity as well. The framework is then applied to a sample of sentences with Need to extracted from the British National Corpus (BNC). The quantitative study shows that Need to is closely connected with discourse-internal sources, conditional sources, and circumstantial sources. It is not closely associated with a subject-oriented source as it has been defined in the taxonomy and in the context of “rules and regulations.” There are significant differences between the spoken and the written modes in the context of conditional sources and discourse-internal sources.

Keywords

source of the modal necessity root necessity verbs

1. Introduction

This paper addresses the concept of “source” of modal necessity (see Depraetere & Verhulst 2008; Depraetere & Reed 2011): it puts forward a principled framework for the analysis of sources of root (or non-epistemic) modality, which is then applied to a sample of examples of need to from the British National Corpus (BNC).¹ In this way, this paper also makes the meaning profile of need to in British English more explicit, by identifying the modal sources that it is typically associated with. Special attention is given to possible differences between the spoken and written mode.

The concept of a source of the (root) necessity refers to who or what makes it necessary for someone to do something or that makes it necessary for a situation to actualize. For instance, in (1), it is clearly the speaker, Ruth’s mother, who tells her daughter what to do and who imposes her will. In (2), it is likewise the speaker who does not want to become involved in the addressee’s plans and who explicitly requests him not to include her in his plan of action. In (3), on the other hand, it is specific circumstances (possibly disagreement among the ruling party members or a situation in which the parties are evenly balanced) that make it necessary for the Government to depend upon minority parties when specific proposals are being put to the vote.

(1) ‘Ruth, you must do as I tell you. Do as I tell you.’ Her mother’s voice, angry, at its wits’ end. (BNC, W_fiction_prose)

(2) You seem to be seeking to destroy yourself in some way, but you must not include me in your plan of action. (Verstraete 2001:1507)

(3) When the parties are evenly balanced in the Commons or when the Government has to depend for support upon minority parties, the Prime Minister may well let it be understood that he will not resign or call an election on an adverse vote in the Commons except on an explicit issue of confidence. (BNC, W_non_ac_ polit_law_edu)

While the role of what is called here “source” of the modality has been treated in discussions of root necessity, previous analyses of need to show that the subjective-objective distinction used in discussions of other modal verbs needs to be complemented: subject-internal necessity needs to be addressed. This is the first topic that will be discussed. Secondly, the paper also sheds light on the semantic profile of need to, by applying the taxonomy of modal sources to a set of examples of need to from the BNC. A better understanding of this facet of its meaning can help capture its typical contexts of use. Indeed, even though need to is a modal verb that is on the rise, it has received comparably less attention than have to, should, ought to, and must.

The paper is structured as follows: in section 2, I position the topic of source of the necessity within previous research concerned with the factors that drive the speaker’s choice of a specific modal verb of necessity from the wide range of available verbs (e.g., have to, must, need to, ought to, should). In section 3, the concept of modal source is explained and a comprehensive taxonomy is presented, with due attention to subject-oriented necessity. The taxonomy is put to the test in section 4: it is applied to a set of examples of need to from the BNC. I describe the data set and method used for the corpus analysis (section 4.1), present the results (section 4.2), and discuss the significant patterns that have emerged from the investigation (section 4.3). The analysis can potentially substantiate or invalidate previous claims about the typical contexts of use of need to, and it can shed additional light on the taxonomic distinctions involved. In the conclusion, I suggest ways in which the notion of source can be exploited in further research.

2. Background

In research on the semantics of modal verbs that express root necessity, various factors have been argued to determine their contexts of use. Analyses of this kind often involve a comparison of pairs of verbs like have to and must or should and ought to. Tagliamonte (2004), for instance, makes use of a 1.2 million word corpus of York English to examine changes in the use of root must, have to, have got, and gotta across age communities. In doing so, she assesses to what extent internal linguistic features like animacy, type of verb (stative, durative, or punctual), and grammatical person shape the contexts of use of the relevant modal necessity verbs. Tagliamonte and Smith (2006) investigate several syntactic and semantic features that possibly differentiate the contexts of use of must, have to, have got to, and got to in various English, Scottish, and Irish communities and also consider the impact of external sociolinguistic factors, such as age and gender (also see, e.g., Larreya 1982; Schützler & Herzky 2021). Myhill (1997) uses data from American plays written since 1889 to get a better view on what differentiates root should from ought to; one of his conclusions is that should is used to express recommendations which are individual opinions, whereas ought to communicates general agreement. Cappelle and De Sutter (2010) examine by far the largest set of (mainly) formal features that possibly impact on the choice of should versus ought to, including properties of the subject (such as animacy and number), the verb (such as situation type and tense), and the clause (such as polarity and embedding).

In this paper, I examine in detail the factor that has been addressed by some literature (e.g., Lyons 1977; Verstraete 2001) under the heading of the subjective versus objective nature of the modal meaning. Verstraete (2001) offers an overview of the (various) meanings that subjectivity/objectivity cover in research on modality (and more in general, within cognitive linguistics). His view is that subjectivity should be defined in terms of modal performativity, meaning that the speaker takes a specific position of “commitment with respect to the propositional content of the utterance” (Verstraete 2001:1517). He uses, amongst others, interrogative contexts as a diagnostic test to assess the subjective/objective nature of the modal. He argues that “subjective modals are oriented towards the interlocutor under the influence of interrogation, whereas objective modals remain unaffected” (Verstraete 2001:1521). For instance, when the declaratives in (4) and (6) are turned into interrogatives, the necessity is oriented towards the addressee in the case of subjective necessity (5), but not in the case of objective necessity (7).

(4) What we want is for the right honourable gentleman to use the full weight of his office. We are getting tired of a cosmetic approach an oversanguine approach. There is a crisis and he must act now. (Verstraete 2001:1521)

(5) Must the minister act now? (Verstraete 2001:1521)

(6) Brake shoes must always be renewed in sets of four. (Verstraete 2001:1521)

(7) Must brake shoes always be renewed in sets of four? (Verstraete 2001:1521)

Whereas “Do you want the minister to act now?” captures the meaning of (4) (the modality is subjective), “Do you want break shoes always to be renewed in sets of four?” does not (the modality is objective).

Even though Kratzer’s (1981, 2010) terminology is different, she also differentiates different kinds of so-called conversational backgrounds in her formal model: they constitute modal basis and serve to define different types of modality. For instance, phrases like given the regulations, in view of what the law provides, and in view of what is commanded capture a deontic modal base; in view of my goals and in view of the circumstances respectively capture a teleological and a circumstantial modal base. Bouletic conversational backgrounds have to do with preferences and wishes. While there are restrictions on the kinds of conversational backgrounds that can be used to interpret modals, they mainly define the modal meaning that is at stake.

In this paper, I take up Verstraete’s (2001) distinction; in doing so, I address, in more detail, what he calls “objective modality” by making use of the concept of source of modal necessity. I also offer a detailed discussion of subject-oriented source, and I show how the different modal sources can be operationalized in empirical data analysis. A more explicit taxonomy enables a finer-grained analysis of the root necessity meaning involved; indeed, on the basis of the empirical study reported on in section 4, I show what conclusions can be drawn about the semantic profile of need to in terms of the sources that it is typically (not) associated with.

3. Sources of the Necessity: An Overview

3.1. Source: A Basic Taxonomy

In this section, I clarify and analyze the semantico-pragmatic feature of source; the different types of source are illustrated with examples of need to taken from the BNC and from the British English component of the International Corpus of English (ICE-GB, Greenbaum & Nelson 1996). The compound adjective “semantico-pragmatic” is meant to reflect the nature of source: it is “semantic” in the sense that every modal verb is associated with a specific source; “pragmatic” indicates that the source is identified in the linguistic and extra-linguistic (that is, discursive and broader situational) context.

Depraetere and Verhulst (2008:3) examine root necessity as expressed by have to and must. Modal source in their analysis is defined as the “driving-force behind the necessity”:

Any necessity (be it expressed by have to or must) is linked up with a source: if there is reference to a necessity, it must originate somewhere, i.e., some entity must be at its origin. This hypothesis is obviously motivated by our knowledge of the world and is as such not concerned with the linguistic expression of sources.

The typology of sources in their paper is based on a qualitative and quantitative corpus study. The major distinction made is that between discourse-internal (subjective) and discourse-external (objective) sources. A discourse-external source may be of three types: the modal source may be a regulation (or a rule), circumstantial, or conditional.

(i) Discourse-internal (speaker/hearer)

(ii) Discourse-external

(a) Rules and regulations

(b) Circumstantial

In what follows, I show that an additional category, that of subject-internal source, needs to be added to arrive at a taxonomy that covers the full range of potential sources of necessity. Subject-internal modality has been primarily associated with root possibility: it is the defining feature of ability. In Jennifer can sing well, the ability of singing is attributed to the skills of the referent of the grammatical subject, which therefore constitutes the source of the possibility. However, this feature also occurs in discussions of modal necessity, in particular in discussions about need to, but it is applied to a very diverse set of examples, so overall, its contours remain quite blurry. As shown in 3.3, it is therefore useful to investigate the notion of subject-internal source in more detail in order to arrive at a more rigorous definition and a principled framework for the analysis of modal sources of necessity overall.

3.2. Discourse-internal Source

In examples with a discourse-internal source, it is either the speaker (in declarative sentences) or the hearer (in interrogatives) who is the origin of the necessity, as illustrated in (8)-(11). The concept is present in early corpus-based descriptions of modal verbs and in reference grammars under the heading of subjective modality (e.g., Coates 1983; Huddleston & Pullum 2002).²

(8) B: He described to me in general the detail his business e.g. where he operated from how many people he had and who the directors were.

A: Yeah but you knew all that didn’t you

B: No I didn’t at that time because it was twelve thirteen fourteen months later

It could’ve all changed

I didn’t know his turnover because the accounts I had were originally nineteen eighty-five were not up to nineteen eighty-eight

A: And uhm so you wanted to see some accounts then (unclear words) first thing you need to do is give us an up to date set of accounts. (ICE-GB: S1B-065, legal cross-examinations, Queen’s bench London)

(9) At the moment there are no savings identified and it seems to me that this Committee needs to deal with the budget as it finds. (BNC, JJ9 S_meeting, environment committee meeting)

(10) I think we need to write three letters. One to the police, one to the ambulance, one to the fire brigade. (BNC, KD8 S_conv)

(11) […] it isn’t particularly clear whether in fact this is part of er some idea that the government has that the people shouldn’t register and I think the minister needs to be very clear about this so that people get the message outside. (BNC, JSG S_Parliament)

The sentence with need to in (8) is directive: the speaker tells the addressee to do something; he is in a position of authority and orders the addressee to hand in accounts. In (9), (10), and (11) as well (which are not directive in the sense that the speaker is giving an order to the addressee), the speaker is the source of the necessity, and it could potentially be argued that the subjective nature is enhanced by markers such as “I think” and “It seems to me.” It is important to add that, even though the speaker’s wish to bring about a situation may be based on specific information that is publicly available and even though she may well be in a position to motivate it on the basis of specific circumstances, in each of the examples, the situation is clearly represented as necessary because the speaker believes it must happen. For instance, in (9), the speaker is expressing her view, taking into account decisions made at previous meetings of the Environment Committee, but this is all the same her personal opinion that she voices when she says that it is necessary for a specific situation to actualize.

In interrogative clauses, it is the addressee who constitutes the discourse-internal source. In example (12), “Do I need to take both packets?” can be paraphrased as “Do you think it is necessary for me to take both packets?” (example 5 is similar).

(12) (SP: PS0E8) There’s your cartridges.

(SP: PS0E9) Do I need to take both packets?

(SP: PS0E8) I wouldn’t think so, I can’t think you’ll get through ten cartridges myself, that seems rather excessive. (BNC, KCD S_conv)

Here again, one might want to argue that there are two sources at stake in this example, in the sense that the hearer is being asked whether there is a (circumstantial) need. In the operationalization of source in this paper, examples of this type are classified as discourse-internal, also taking into account Verstraete’s (2001) observations (see section 2). The paraphrase is clearly “do you believe it is necessary?” In other words, the hearer is the primary source of the necessity. She can potentially pick one from a range of reasons why something is (not) necessary, specific circumstances being one of them, but the hearer is clearly called upon as the primary decision maker.

Given the fact that the term “subjective” covers quite diverse notions in linguistics in general, the label “discourse-internal” is used in this paper to avoid terminological confusion. This label also nicely captures the fact that it can be the speaker or the addressee who is the source of the necessity, depending on the sentence type in which the modal verb is used.

3.3. Subject-internal Source

As pointed out in section 3.1, the basic taxonomy needs to be supplemented with a source which more commonly features in accounts of root possibility, more specifically in discussions of “dynamic” modality (e.g., Palmer 1990; Huddleston & Pullum 2002). Volition (e.g., 13) and ability (e.g., 14) are typical examples of dynamic modality in taxonomies of modal meaning; it is modality “concerned with properties and dispositions of persons, etc., referred to in the clause, especially by the subject NP” (Huddleston & Pullum 2002:178). The term “subject-internal” source is used to capture the kind of modal source in cases like these. In (13) and (14), the modal source is located within the referent of the grammatical subject, who appears as the agent in control of the situation referred to.

(13) I hope that the House will support the motion. That would help to show the universal determination of all parties in the House to achieve an answer that will provide a proper future for British agriculture and enable me at least to be able to tell my companions in the negotiations that there are things that matter so much to us that we will sit there until we achieve them, even if it means that we will sit there for many more months to come. (BNC, HHX_W_Hansard)

(14) And though I can swim pretty well, the swirl of the tide was fierce, and looked dangerous; and besides, I had no desire to swim. (BNC, CKF_W_fict_prose)

In the case of modal verbs of necessity as well, there may be cases in which the referent of the (grammatical) subject (be it first person, second person, or third person) feels that it is necessary to bring about a certain situation. Examples like these (such as 15 and 16), in which the subject referent intentionally wants to do something (witness I want to tell you, I would much prefer to tell you in 15), are therefore classified as having subject-internal sources.

(15) […] I miss Paris and I miss you! We never spoke much, though that doesn’t really matter as words are often inappropriate when in the presence of feeling. Having said that there are one or two things I want to tell you, and for the moment the only way I can express them is through this letter. I would much prefer to tell you one-to-one. Anyway, prepare yourself. You are special, because you are you. You are lovable, and I … You are unique, and talented too. Are you. You probably didn’t expect me to say such things Francoise, but I needed to say them.

I could say more, a part of me wants to, but there’s no need for me to embroider or elaborate on my words, that would ruin the genuineness of my feelings. I will change the subject now and tell you something different. (ICE-GB W1B-008)

(16) You feel though you’ve got to come out, you need to get out! You know, he and he’s at that point now … Mm where he needs to get out. Which is fair enough. (BNC, KP1_S_conv)

In (16), there is reference to a person who is at the point of leaving a hospital; the paraphrase of the sentence with need to is ‘He feels it is necessary for him to get out.’ In both examples, volition is involved, and the “order” is addressed by the subject-referent to the subject-referent as it were. Modal necessity verbs in this use are close in meaning to want to.

Subject-internal necessity has received a lot of attention in discussions about need to, under various headings, as this modal verb is considered to be typically associated with this type of source. A variety of labels has been used (e.g., “participant-internal,” “internal compulsion,” “internally motivated,” “irresistible inward urge,” “irresistible compulsion”). The overview that follows will show that this concept has fuzzy borderlines and that the more rigorous approach that is outlined here, whereby it is embedded in a full-fledged taxonomy and systematic framework for the analysis modal sources, is helpful.

For Nokkonen (2006), the basic feature of subject-internal necessity is that it is weak (in force) and objective. In other words, it is not the modal source as such that is the only criterion for assigning a sentence to this category on her approach. For instance, examples (17) and (18) are said to belong to the “internal group” because they are objective and weak.

(17) A no, wait a minute

B yeah

A no, I mean, you’d get to

c well, you need to get to Putney, Hammersmith

A then you met Ham, you you hit Hammersmith, and you’ve got to get on from there, that’s all right. (Adapted from Nokkonen 2006: 45)

(18) I need to know I need to so keep me informed, if you really feel that things are that bad that you need to do it, [Yeah.]

[then] consult me, okay. (Adapted fromNokkonen 2006: 45)

Nokkonen (2006, 2010) starts from Van der Auwera and Plungian’s notion (1998) of “participant-internal necessity,” which refers to “necessity internal to a participant engaged in the state of affairs” (1998:80); they give the necessity example seen in (19) to illustrate this type of modal meaning.

(19) Boris needs to sleep ten hours every night for him to function properly. (Van der Auwera and Plungian, 1998:80)

In Van der Auwera and Plungian’s (1998) paper, participant-internal necessity is the necessity counterpart of participant-internal possibility, which is used to characterize typical “ability” examples (e.g., Boris can get by with sleeping five hours a night). The examples from Nokkonen (2006, 2010) show that Van der Auwera and Plungian’s (1998) original concept is applied to a far wider range of cases than intended by Van der Auwera and Plungian (1998). The instances in (20) and (21) are further examples of this category, which Nokkonen alternatively calls “internal compulsion” (Nokkonen 2006) and “participant internal need to” (Nokkonen 2010).

(20) I need to go now. (Nokkonen 2006:37)

(21) Ann: I’ve wiped this I’ll have washed the whole floor.

Sally: <laugh>

Ann: I won’t need to do it tonight. I’ll get some knitting done. Seeing as I haven’t any ironing to do. (Nokkonen 2010:48)

The examples do not appear to be entirely identical in nature when considered from the point of view of source as it has been defined here. With regard to (20), Nokkonen (2006:37) points out that the subject referent feels a compulsion which originates within herself: “such compulsions, even if the subject is first person I, are objective, since the speaker has no conscious control over them” (I discuss this observation in more detail in 3.4.2). However, without further context, it seems that example (20) could also be a case in which conscious volition is involved or one in which the necessity to leave is motivated by a circumstantial source that lies outside the subject. It is difficult to identify a compulsion or irresistible urge without access to the wider context.³ Likewise, in (21), it is the circumstance of the floor having been cleaned that does not require any further action. In other words, these examples show that subject-internal has become a category that is far broader in scope than Van der Auwera and Plungian’s (1998) participant-internal modality on which it is based.

Declerck (1991:386) is another author who brings up subject-internal necessity in his discussion of need to. He uses the term “internal necessity”, which is defined as follows: “the idea that something is required for personal reasons […] this is a necessity whose origin lies in the referent of the S [Subject] himself, or in his own situation.” He gives the examples and paraphrases in (22)-(24) for illustration.

(22) She needs to take pills every day. (Her state of health makes it necessary for her to take pills every day.)

She has to take pills every day. (She is obliged to.)

(23) You need to have a bath. (For your own sake that I’m saying this: you look untidy.)

You must have a bath. (I want you to.)

(24) He needs to talk to others about his work. (because of his character)

(25) Children need to learn to respect the property of others. (because it is an essential aspect of their education)

While the absence of a wider context makes it harder to interpret the examples, they could be described as follows in terms of modal source of necessity as defined in this paper: in (22), the health condition of the subject referent requires her to take pills; this example illustrates what I call a circumstantial necessity (section 3.4.2). In (23), the paraphrase in brackets pinpoints two facts: (a) the subject referent is the beneficiary (who benefits from bringing about the necessary situation is a different criterion that is not addressed in this paper); (b) the speaker believes that it is necessary for the addressee to have a bath. In other words, the source is discourse-internal. In (24), the subject referent seems to be subjected to an “irresistible urge.” This example illustrates what I call a (circumstantial) subject-driven source (section 3.4.2). Example (25) illustrates “rules and regulations,” as customs or good practice also belong to the realm of binding forces (section 3.4.1).

Declerck (1991:382) uses the phrasing “inner compulsion” and “irresistible inward urge” when commenting on examples with accented must, like those in (26), and observes that here, must is almost interchangeable with will, expressing “obstinate insistence.”

(26) I simply MUST have that necklace. I will not go home without it.

As the subject referent is the agent in control of the situation (and potentially very eager to bring it about) rather than the patient subjected to an inner compulsion (“subject-driven”), my view is that this example is the only one in (22) to (26) that illustrates a subject-internal source.

Westney (1995) and Collins (2009) also associate need to with a necessity that is in some way “internal” to the subject referent (or to the speaker). Westney (1995:137) uses “internally motivated need” when commenting on the example given in (27).

(27) The prosecutors, Marcia Clark and Christopher Darden from the district attorney’s office, described the defendant as a violent wife-beater who needed to control Nicole Simpson in every way.

Westney (1995:137) writes: “This refers to an internally motivated need and suggests that need (to) is in fact vague as the source of motivation it signals. […] There is some similarity between the use of need (to) for internally motivated need seen here and the use of have to (and to some extent, must) illustrated in (4.145)-(4.147).”I cite here his examples (145)-(147), as (28)-(30), for illustration.

(28) When we ate cherries, we didn’t just try and suck them in by the stalk, oh no, we had to go one better. (Westney 1995:135)

(29) So he said, right well, you’d better ring me then so I said, OK, well, I’ll ring you tonight some time, so I told the others about it and we all had to have hysterics again. (Westney 1995:135)

(30) The rain was bad enough, but then it had to snow. (Westney 1995:136)

In (28) and (29), the subject’s behavior is said to be presented as “motivated by some inner, irresistible compulsion” (Westney 1995:136), which is illustrated in Quirk, Greenbaum, Leech, and Svartik's (1985:226) example in (30) too. It will be clear that the “vagueness” that Westney (1995) refers to in his comment on the example in (27) relates to the trigger of the (irresistible) internal urge, and does not in itself question the presence of an irresistible compulsion.

Collins (2009:73) offers a view similar to Westney’s (1995). He argues that in sentences with a first person subject, like I need to get ready, need to suggests “compulsion arising from within the speaker. It is from this sense of internal compulsion that the deontic meaning of need to derives. Statements pertaining to an addressee’s needs come—via indirect illocutionary force—to serve the role of recommendation or exhortations.”⁴ Even though Collins (2009) first associates internal compulsion with the referent of the speaker (rather than with the subject referent), he also mentions sentences (with a second person subject) in which the needs relate to the subject referent. He agrees with Smith (2003: 260) (see Nokkonen 2006:37) that need to “can acquire the force of an imposed obligation, but—something which does not apply to the other markers—the writer or speaker can claim that the required action is merely being recommended for the doer’s own sake.” In other words, the situation to be actualized is presented as beneficial to the addressee (cf. also Müller 2008:82).

This overview shows that subject-internal source is not understood in the same way by the authors cited here, and it follows that the views on the nature of need to, despite the use of similar terms, diverge and are even to some degree incompatible. This observation is relevant to the discussion of the data analysis in section 4. The definition argued for here is that a subject-internal source is one in which the modal source is located within the referent of the grammatical subject, who controls the situation referred to. Agentivity, in the sense of Cruse’s (1973:18) “volitive” (i.e., “is present when an act of will is stated or implied”), is taken as the defining criterion for subject-internal source in sentences with an animate subject referent.

3.4. Discourse-external Sources: Rules and Regulations, Circumstantial Sources, and Conditional Sources

3.4.1. Regulations or Rules

The category of regulations or rules (as in 31 and 32) is relatively straightforward: in many cases, the necessity for a specific situation to be brought about resides in an existing rule or a regulation.

(31) I have your name on a list of people who need to do a 4 day First Aid course because you take students on field trips. (ICE-GB, W1B-018)

(32) For an article in an anthology, for instance, you need to keep details of the journal in which it originally appeared. (BNC, HXH W_ac_humanities_arts)

In (31), there is reference to a regulation that requires people who take students on field trips to follow a first aid course. In (32), the style guide of anthologies requires journals to be mentioned in a specific form. Examples (33) and (34) are further illustrations of this category.

(33) If, for example, you asked someone not to smoke in your presence (the right to ask others to respond to your needs and wants) you need to respect their right to have different needs and wants and, much more difficult, their right to say no. (BNC, B2F, W_commerce)

(34) right now let’s see what we need to do (pause) it says (reading) we need to melt (unclear) the margarine. (BNC, KBW S_conv)

Rules and regulations also encompass, for instance, assumptions about what is considered morally appropriate behavior (33) and cooking instructions (34).

3.4.2. Circumstantial Sources

Examples that refer to a specific (temporary or habitual) situation which necessitates the actualization of a situation are classified here as having a circumstantial source. In (35), for instance, the situation requires large quantities of photocopies to be made; in (36), it is the specific situation of traveling which urges the backpacker to split the weight of the tent she’s carrying. In (37), it the nature of words, potentially all words, that necessitates the use of quotes in a specific way.

(35) Any photocopying that needs to be done in great bulk, erm get quotes, and let the company pay for it rather than you paying for it and then pay you back. (BNC, F8U S_meeting)

(36) Wetter climates favour tents that go up fly first so that the inner and your gear can be protected as quickly as possible. Pitching inner and fly together can be a time saver, but obviously the inner must be detached from the fly if you need to split the weight for backpacking. (BNC, CHJ W_ pop_lore)

(37) Philosophically, it could be argued that most or all words present analogous difficulties, and need to be put in scare quotes (“the ‘topic’ of this ‘essay’ is the ‘issue’ of the ‘author’s’ ‘intentions’”, etc.). (BNC, HXH W_ac_humanities_arts)

I have also included, as a circumstantial source, examples in which the subject referent appears as a patient subjected to an inner urge beyond her control. In this case, the modality may be said to be subject-driven in the sense that the subject referent is under the compulsion to do something and therefore, the modal source can be considered as a kind of circumstantial source. Example (38) is a case in point: the speaker’s body signals a need to go to the toilet, a need which is beyond her control.

(38) Oh, I need to go to the toilet, hold on. (BNC, KDN_S_conv)

Example (39), in which want to and need to are contrasted, illustrates the difference between a subject-driven (circumstantial) versus subject-internal source.

(39) What does the circle mean in this play? What does it represent? What is the “Keeper” looking after? Who is the “Keeper”? What does the newspaper represent? Who, or what, is B, the thief? And why does the thief want or need to take the “keys”? Providing the teacher retains an open mind, an extraordinary drama can evolve from this. (BNC, HYA W_ac_ polit_law_edu)

Here, need to expresses circumstantial necessity: the thief is required to do something, either as a result of an irresistible character trait (which is called here subject-driven necessity, a subtype of circumstantial necessity) or possibly because there is a specific situation, not specified in this example, which required him to do so. Subject-internal necessity originating in the subject referent’s conscious volition is expressed by want to in this example.⁵

3.4.3. Conditional Sources

In the case of a conditional source, the actualization of a situation is represented as necessary in order for another situation to actualize. The desired situation may be literally mentioned in the context, in the form of a subclause of purpose or result or the protasis of a conditional clause. Examples are given in (40)-(42).

(40) In order for strategic planning and decision making to be effective, a different set of values needs to be held by senior officials. (BNC, GVN W_commerce)

(41) Witness the image of the watch: it remained easy to think of the system as having a grand purpose built into it. To explain a watch we need to realize both how its springs and wheels (or, nowadays, microchips) work and that its purpose is to tell the time. (BNC, EDD W_ac_polit_law_edu)

(42) It is the kind of approach the council needs to take if we are to avoid the kind of apartheid against disabled people which this town may develop. (BNC, K55 W_newsp_other_report)

All of the examples can be paraphrased as “X is necessary in order for Y to actualize” or “X needs to actualize in order for Y to actualize.” In (40), it is necessary for senior officers to hold a specific set of values (X) in order for strategic planning to be effective (Y); in (41), having insight into how the springs and wheels of a watch work (X) is represented as a skill that is required in order to explain how a watch works (Y). Finally, in (42), it is necessary for the council to take a specific approach (X) in order to avoid apartheid against disabled people (Y). In other words, what sets these examples apart from circumstantial examples is that there is a relationship of posteriority between the situation that needs to actualize (at time t) and the required result to be produced (at time t+1). Another observation is that, in the majority of conditional examples, the situations that are presented in the if-clause or in the subclause of purpose are perceived as positive.⁶

The aim may also be clear from the surrounding linguistic context, as seen in (43). Here, need to is said to have a conditional source because the sentence can be paraphrased as “You need to have certain skills in order to be an effective coach.”

(43) An effective coach is interested in people, recognizing their importance and potential. You need to know the desires, interests and capacities of your subordinates and above all to avoid doing their thinking for them. (BNC, AYJ W_commerce)

Example (44) is interesting as it potentially involves two sources.

(44) In order to qualify for a full pension, you need to have paid into it for over forty years. (BNC, FX5 S_brdcast_discussn)

There is reference in (44) to existing pension regulations and there is also a subclause of purpose (X = you need to pay for forty years, Y = you qualify for a full pension). Examples like this are categorized as conditional: even though the sentence is part of the set of rules concerning pensions, it is not the rule as such that imposes the actualization of a situation. Rather, having paid into a pension is presented as being necessary in order to qualify for a full pension. Again, we observe the temporal relationship of posteriority between the proposition in the main clause and that in the subclause of purpose.

3.4.4. Interim Conclusion

Discourse-external sources have been spelled out in terms of three categories. If “circumstances” is defined broadly, it might be argued that rules and regulations and conditional sources are also circumstantial. However, here, “circumstantial” is understood in more narrow way, as (a) circumstances that are not rules and (b) circumstances that are not posterior situations whose actualization is explicitly presented as an aim.

The observations made here are explored in more detail in the data analysis in section 4. I show that there is a contradiction between the claims that have been made in the literature about the distinctly subject-internal nature of need to and the scant evidence for this claim in the BNC dataset examined in this paper. The distinctions made here are therefore be taken up again and discussed in further detail in section 4, where I apply the taxonomy to a sample of examples of need to from the BNC.

4. Modal Source and Need to

With the notion of source and its relevance to the semantico-pragmatic profile of modal verbs established, I show how this notion can be exploited in data-analysis, making use of a sample of need to examples. My choice to investigate need to is motivated by changes in the frequency of modals that have been reported in the literature: investigations into (recent) diachronic change have shown that need to is claiming an increasingly important space in the paradigm of modal verbs expressing obligation, both in American English and in British English (Leech 2003, 2013; Leech et al., 2009; Mair 2015; Mair & Leech 2021; Smith 2003).⁷ Here I zoom in on the meaning of need to from the perspective of modal source and in this way shed new light on the semantic profile of a modal verb that is on the rise. Smith (2003), Taeymans (2004), Nokkonen (2006, 2010), Kastrone (2008), Müller (2008), and Van der Auwera and Taeymans (2009) discuss various facets of the contexts of use of (mainly) the lexical modal verb need to. Müller (2008) looks at the diachronic development of need to; register (Kastrone 2008) and sociolinguistic variation such as gender and age (Kastrone 2008; Nokkonen 2010) have also been addressed. Van der Auwera et al (2012) examine regional variation, more specifically the use of need to in Asian varieties of English compared to that in American English and British English. However, while these empirical studies investigate the pragmatic and semantic profile of need (to) to a certain degree, they are primarily focused on formal factors (such as passive infinitive versus active infinitive, person and number of the subject, and negative versus positive polarity contexts). Compared to studies of the other necessity modals, analyses of need to devote much less attention to the modal source. It is this facet of the semantic profile of need to that the present paper addresses.

4.1. Material and Method

My data set consists of 585 tokens of need to extracted from the BNC,⁸ accessed through the online interface created by Mark Davies (2004). First, I determined the relative frequency (per million words) of the different forms of need to (base form indicative present, base form infinitive, third person singular indicative present, preterite) in the written BNC subcorpus (90 million words) and the spoken BNC subcorpus (10 million words). The progressive form (is/was/are/were needing to) was not included because this form is very infrequent (5 tokens). I then extracted a random sample of 585 tokens (as the relative frequency of the lemma is 585 pmw) whose distribution mirrors that of the distribution of need to in the whole BNC.⁹ For instance, NEED to is used proportionally more often in spoken English (see Kastrone 2008:74 for a similar finding); accordingly, the spoken examples make up 66 percent of the data set, which corresponds to the distribution of need to across the spoken-written modes in the overall corpus. The composition of the sample is shown in Table 1.

Table 1.

Frequency of need to Forms

Mode	Need to (indicative)	Need to (infinitive)	Needs to	Needed to	Total
Spoken	222	107	45	12	386
Written	81	55	40	23	199
Total	303	162	85	35	585

This data set was coded in terms of the source of the modality for each example. The annotation of a semantico-pragmatic feature like modal source is challenging. However, collaborative work on modal source for over ten years (Depraetere & Verhulst 2008; Depraetere & Reed 2011; Depraetere et al., 2021; Verhulst et al., 2013) has enabled me to fine-tune the parameter and to make it fully operational for data analysis. It was therefore always possible to identify the source and classify the examples in the data set presented here. In this kind of analysis, it is important to consider the wider linguistic context as well as the general discourse context. I analyzed the data set three times, with intervals of several months between the different rounds of analysis.

I looked for two kinds of significant patterns: two chi-squared tests were run in R, one for the overall distribution (that is, to check if the observed distribution of sources significantly diverges from an even distribution), and one to see whether there are significant differences in the distribution of sources between the spoken and the written modes.¹⁰

4.2. Results

Table 2 gives an overview of the distribution of sources in the sample analyses. The overall distribution is statistically significant (χ² = 185.97, df = 4, p-value < 2.2e-16). That is, given that there are five sources, the assumption is that they will be distributed evenly in the randomly selected dataset. But the distribution reported in Table 2 is such that the observed distribution significantly diverges from an even distribution. The residuals mentioned in Table 2 are values that show what each cell contributed to the overall chi-squared value. When the value is larger than +2/-2, the individual difference is significant (Meyer et al., 2003:4). As seen in Table 2, the residuals for each source are larger than +2/-2, and they consequently all significantly differ from the even distribution: discourse-internal, conditional, circumstantial sources occur more frequently than expected, and rules and regulations and subject-internal sources appear less frequently than expected. In other words, the distribution of each of the sources contributes to the overall chi-squared value.

Table 2.

Overall Distribution of Sources

Source	N	%	Residuals
Discourse-internal	175	29.9	5.084752
Conditional	172	29.4	5.362102
Circumstantial	160	27.4	3.975351
Rules and regulations	65	11.1	-4.807402
Subject-internal	13	2.2	-9.614803

Table 3 shows distribution of sources across spoken and written modes.

Table 3.

Distribution of Sources Across Modes

Mode	Discourse-internal (%)	Conditional (%)	Circumstantial (%)	Rules and Regulations (%)	Subject-internal (%)	Total (%)
Spoken	150 (38.9)	67 (17.4)	111 (28.8)	50 (12.9)	8 (2)	386 (100)
Written	25 (12.6)	105 (52.8)	49 (24.6)	15 (7.5)	5 (2.5)	199 (100)
Total	175 (29.9)	172 (29.4)	160 (27.4)	65 (11.1)	13 (2.2)	585 (100)

With respect to distribution of sources across the spoken and the written modes, the distribution is statistically significant (χ² = 90.74, df = 4, p-value < 2.2e-1). Residuals make it possible to pinpoint where the significant differences in distribution lie; it is possible to represent the relative contribution of each cell to the overall chi-squared value in an extended association plot on the basis of residuals. The extended association plot in Figure 1 visualizes, on the basis of residuals that exceed +2/-2, which of the cells in Table 3 significantly contribute to the significant result of the χ², that is, which sources are significantly associated with one or the other mode. The dark grey tiles show significant associations: those that extend above the line represent a significant positive association (attraction) with a mode, while those below the line represent a significant negative association (repulsion) with a mode. The light grey tiles represent non-significant associations.

Figure 1.

Significant Patterns in the Distribution of Sources Across Spoken/Written Modes

The following significant patterns stand out: conditional sources typically occur in the written mode and typically not in the spoken mode; discourse-internal sources typically occur in the spoken mode and typically not in the written mode.

4.3. Modal Source and Need to: Discussion

4.3.1. Discourse-internal Source (Speaker/Hearer)

Need to typically features in contexts with a discourse-internal source; almost one out of three examples (29.9 percent) belongs to this category. The source is discourse-internal in (45) and (46), for instance: it is clearly the speaker (in 45) or the hearer (in 46) who is the source of the necessity.

(45) (SP:PS3JJ) All I’d say on five chairman is chairman is that this issue of infrastructure particularly sewerage, sewerage, foul water and water supply is a major issue in York and the location of the new settlement I think needs to be very closely assessed in terms of er of five. I think it’s a very important criteria that we need to look at.

(SP:PS3JK) I agree it’s important, what I have no feel for, is whether it is something which helps to make a decision as to where the new settlement might go. What I think I’m looking for is some evidence that the severity of the infrastructure problems to which you refer actually varies from one district to another. (BNC, HVH S_ pub_debate)

(46) (SP:PS6G8) Are we having (unclear) we have to vote here? Anything else needs to be (unclear). Anything you need to raise with us?

(SP:PS6G8) No, I don’t think so. I mean the main thing is these new leaflets really

(SP:KRYPSUNK) mhm.

(SP:PS6G8) getting them produced, checked

(SP:KRYPSUNK) You’re happy about that?

(SP:PS6G8) Yes. (BNC, KRY S_meeting)

This generalization is in line with Larreya and Rivière’s (2019:128) description: they argue that need to often expresses a “subjective obligation,” and therefore resembles must, the difference being that must is more formal. Translated into the taxonomy used here, their view is that need to is frequently discourse-internal (see 3.2). In (47), from Larreya and Rivière’s (2019), it is the speaker who tells Sophie how she should behave.

(47) Sophie, you need to respect other people’s privacy. (D. Brown, The Da Vinci Code) (Larreya & Rivière 2019:128)

Figure 1 shows that NEED to is typically associated with a discourse-internal source in the spoken mode (150 out of 175 examples) and typically not in the written mode (25 out of 175 examples).

In his analysis of obligation modals, Smith (2003) discusses what he calls “private obligation,” which seems relevant to the discussion of discourse-internal sources. He quotes the examples given in (48)-(51) from Sweetser (1990:53-54).

(48) You have to stay home because I say so. (Smith 2003: 244)

(49) ? You need to stay home because I say so. (Smith 2003: 244)

(50) I need to stay home tonight to study for the test. (Smith 2003: 245)

(51) I have to stay home tonight to study for the test. (Smith 2003: 245)

While one may want to argue about whether or not the example in (49) is truly questionable, Smith (2003:244) writes that (49) is not felicitous because need to requires the obligation to be internally motivated, in the sense that the situation needs to be actualized for “the ‘agent’s or doer’s’ own sake.” Smith (2003:245) calls this a “private obligation.” There is a “conflict in the motivational sources”: need to requires reference to private obligation, whereas because I say so makes it clear that there is an “external motivation.” At first sight, Smith’s (2003) claim seems to challenge our findings in the sense that the example suggests that a discourse-internal source is incompatible with the use of need to. However, the challenge needs to be qualified: first, it is not clear if private obligation signals a discourse-internal source (the speaker in declarative sentences), or, rather, a subject-oriented one (the referent of the grammatical subject). Secondly, for Smith (2003), because I say so does not, so it seems, in the first place say something about the source of the necessity (who feels actualization is necessary?), but, rather, about the beneficiary of the situation (actualization will be beneficial to the addressee; cf. actualization for “the ‘agent’s or doer’s’ own sake”; Smith 2003:244). The notion of source as it is used here leaves it vague who will benefit from the actualization of the situation. While it could be interesting to investigate the link between discourse-internal source and beneficiary and its relevance to the use of need to, this is beyond the scope of this paper.

4.3.2. Conditional Source

In (52) and (53), Need to is associated with what has been defined as a conditional source in this paper.

(52) Kirov was a rat caught in a trap of his own making, and time was running out. He needed to summon all the devious cunning with which nature had endowed him if he were to escape. (BNC, CDA_W_fict_prose)

(53) But SQL server alone will not address all the needs of enterprise-wide client server. It needs to evolve in several ways to meet these needs. (BNC, HEP_S_Speech_unscripted)

Both examples represent the actualization of a particular situation (the main clause situation) as necessary and sufficient for the actualization of another situation, which is referred to in the if-clause (52) or the subclause of purpose (53). In other words, the desired situations are the driving forces behind the necessities, whose actualization require the main clause situation to come about: Kirov being smart is necessary in order to escape (52), and the SQL server evolving is necessary in order to the meet the needs (53). Example (53) occurs in the context of a seminar about the implementation of IT architecture: there is reference to a more extensive use of facilities which will require a change in the architecture.

Conditional sources have received relatively little attention in the literature (e.g., Depraetere & Verhulst 2010:8-11; Furmaniak 2020), but grammars, such as Declerck (1991:386) and Dekeyser et al. (1999), occasionally point out that need to in affirmative sentences “mostly emphasizes what is required for a specific purpose” (Declerck 1991:111). Declerck (1991) gives the example in (54) to motivate his claim.

(54) He needs to earn more money if he is to buy a car. (Declerck 1991:386)

My analysis provides support for this claim. Need to is closely associated with a conditional source (29.4 percent). In terms of distribution across modes, need to is closely associated with a conditional source in the written mode (105 out of 172 examples) and typically not in the spoken mode (67 out of 172 examples).

So while examples with a conditional source have not received much attention in discussions of root necessity Need to, this is one of its typical contexts of use.

4.3.3. Circumstantial Source

Another significant pattern is the use of need to in contexts with a circumstantial source (160 out of 585 examples, 27.4 percent; see 3.4.2). There is a relatively even distribution across the spoken and the written modes. There is no significant difference between the modes: need to is not more typically associated with a circumstantial source in spoken British English than in written British English.

Examples (55) and (56) illustrate circumstantial modal sources.

(55) Then I was able to help when young Mrs Thwaites at High Birk Hatt, our nearest neigbours, suffered terribly from toothache and needed to go to the nearest dentist, who was in Barnard Castle. (BNC, W_biography)

(56) You get much above about fifty people and you perhaps do need a some technical assistance with the power. So the power needs to be just slightly slightly raised okay. (BNC, JSA_S_unclassified)

In (55), it is the circumstance of having toothache which makes it necessary for Mrs. Thwaites to see a dentist. The sentence in (56) features in a discussion about the power of voice: if there is a certain number of people present, the power of the voice needs to be adjusted. In other words, specific circumstances make it necessary to raise the power of the voice.

4.3.4. Rules and Regulations

Need to is typically not associated with rules and regulations (65 out of 585 examples, 11 percent). Examples (57) and (58) illustrate this type of source.

(57) So you have built an item of farm equipment that helps you do your job better, then let us know. It doesn’t have to be newly built, though we are limiting entries to machines built in the past five years. Nor does it need to be big or expensive looking. What we are looking for is evidence of imagination, lateral thinking and the FBLdesigner’s ability to solve technical problems. (BNC, ACR_W_ pop_lore)

(58) The trials revealed a number of essential conditions that need to be met by all employers, assessment centres and candidates participating in the occupational competence assessment. Employers need to be committed to providing candidates with access to and use of workplace evidence; opportunities for additional experience; testimonies of their performance; counselling and support; and, ideally, a development programme for achieving the required competencies. (BNC: GXF W_non_ac)

In (57), the entries submitted to a contest need to meet certain requirements; “the items of farm equipment being big or expensive looking” is not one of them. In (58), it is the requirements to participate in the assessment that constitute the source of the necessity.

No observations have been made in previous analyses of need to with respect to the potential impact of rules and regulations contexts on the choice of need to. The data analysis shows that this is not without reason, as it is not one of the most common contexts in which need to is used. Rather, it seems that, among necessity modals, must is more typically used in the context of laws, and rules and regulations (e.g., Collins 2009:35).

4.3.5. Subject-Internal Source

There are thirteen examples of need to with a subject-internal source (see section 3.3) in the BNC data set. In (59), the subject-internal necessity can be paraphrased as “this is the only information that I consider necessary for me to have access to,” or, “this is all I want to know.”

(59) She pours withering scorn upon such ‘painted poupes’ as Joan Collins and Liz Taylor for the hours they devote to their images, as well as the men they choose as partners, but one look at the deep lines of bitterness etched into the faces of those women who renounce loving relationships in favour of cats, gardens or anything else tells me all I need to know. (BNC, G2V_W_ pop_lore)

The data confirm Müller’s (2008) findings based on corpus data from the ARCHER corpus and the Oxford English Dictionary from 1800 to the present. She shows that examples that express participant-internal necessity are rare (Müller (2008) uses Van der Auwera and Plungian’s (1998) concept (see section 3.3), which is called a subject-internal source in this paper). Only 2 out of the 428 sentences she analyzed illustrate this use, compared to the equally low number of 13 out of 585 examples (2.2 percent) in the data set analyzed here.

This type of example may well be similar to the “performative” use of must discussed by Coates (1983), and illustrated in (60).

(60) “You must play this ten times over, Miss Jarova would say” (Coates 1983:33, 38).

While examples like this, and subject-oriented uses of need to as in (59), are likely to those that are chosen by native speakers when they are asked to give an example of a root use of must and need to respectively, they are by no means predominant in actual usage. In other words, they constitute more of a psychological stereotype than a “quantitative stereotype” (Coates 1983:38). Müller (2008) explains the low number as follows: there are few participant-internal examples as this meaning is still expressed by the source form, lexical verb need, from which need to developed. In this sense, Müller’s (2008) and my own findings do not confirm Nokkonen’s (2006, 2010:50) conclusion that “[t]he basic internal meaning is, […], still very important: roughly half of the instances are in this domain.” As shown in section 3.3, the interpretation of the concept of “participant-internal” clearly varies across authors, which makes a comparison challenging: internal meaning, internally motivated need, and inner, irresistible compulsion are associated with different types of examples in previous work. As it has been defined in this paper, subject-internal refers to a necessity whose source lies with the subject-referent and is controlled by her (volition). The data analysis shows that the commonly-voiced view in the literature needs to be adjusted: a subject-internal source is not one of its main defining features.

5. Conclusion

The aim of the paper was to offer a more comprehensive and more explicit taxonomy of modal sources of root necessity. I have shown that the detailed investigation of the modal sources associated with a modal verb can shed new light on its semantico-pragmatic profile, need to having served as a test case in this paper. The findings can be summarized as follows: on a theoretical level, I have developed a principled framework for the analysis of sources of root modal necessity, paying particular attention to the positioning of the concept of subject-internal source (and subject-driven source) to the subclasses of discourse-external sources. From the point of view of usage, it has been shown that in the dataset extracted from BNC, need to is typically used with circumstantial, conditional, and discourse-internal sources and typically not with a subject-internal source or in the context of rules and regulations. In other words, examples with a subject-internal source, while featuring quite prominently in the literature, do not occur very often. The following significant patterns with respect to mode (spoken versus written) have been observed: Need to typically patterns with a discourse-internal source in spoken discourse and typically not in written discourse; Need to appears typically with a conditional source in written discourse and typically not in spoken discourse.

More generally, the method presented in this paper can be useful to arrive at a deeper understanding of the factors that drive the choice of modal verbs of root necessity. Further multi-factorial analyses that include semantic factors (such as the type of verb phrase), formal features (such as the type of subject), lexical features (co-text), and a larger range of verbs will be useful to complement this investigation and to determine the relative role source plays in the selection of a specific modal necessity verb.

Footnotes

Acknowledgements

I am grateful to Benoît Leclercq and Samantha Laporte for useful discussion of various topics addressed in the paper, and to the reviewers and editors for their critical observations and constructive suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the ANR (Agence Nationale de Recherche) (ANR-16-CE93-0009).

ORCID iD

Ilse Depraetere

Notes

Author Biography

Ilse Depraetere is Professor of English Linguistics at the University of Lille. Most of her publications relate to the verb phrase in English (tense, aspect, modals), with a particular concern for the semantics-pragmatics divide. She has written a grammar of English, Advanced English grammar – a linguistic approach, with Chad Langford (2019, 2nd edn., Bloomsbury), and she has recently co-authored papers that analyze linguistic (in)directness and politeness in online complaints.

References

Davies

Mark

. 2004. British National Corpus (from Oxford University Press). https://www.english-corpora.org/bnc/.

Greenbaum

Sidney

Nelson

Gerald

. 1996. The International Corpus of English, British English Component (ICE-GB).

Auwera

Johan van der

Plungian

Vladimir

. 1998. Modality’s semantic map. Linguistic Typology 2(1). 79-124.

Auwera

Johan van der

Taeymans

Martine

. 2009. The need modals and their polarity. In Bowen

Rhonwen

Mobärg

Mats

Ohlander

Sölve

(eds.), Corpora and discourse – and stuff. Papers in Honour of Karin Aijmer, 317-326. Gothenburg: Gothenburg University.

Auwera

Johan van der

Noël

Dirk

De Wit

Astrid

. 2012. The diverging need (to)’s of Asian Englishes. In Hundt

Marianne

Gut

Ulrike

(eds.), Mapping unity and diversity of New Englishes World-Wide, 55-76. Amsterdam: John Benjamins.

Cappelle

Bert

De Sutter

Gert

. 2010. Should vs. ought to. In Cappelle

Bert

Wada

Naoki

(eds.), Distinctions in English grammar. Offered to Renaat Declerck, 92-126. Tokyo: Kaitakusha.

Coates

Jennifer

. 1983. The semantics of modal auxiliaries. London: Croom Helm.

Collins

Peter

. 2009. Modals and quasi-modals in English. Amsterdam: Rodopi.

Cruse

Alan

. 1973. Some thoughts on agentivity. Journal of Linguistics 9(1). 11-23.

10.

Declerck

Renaat

. 1991. A comprehensive grammar of the English language. Tokyo: Kaitakusho.

11.

Dekeyser

Xavier

Devriendt

Betty

Tops

Guy A. J.

Geukens

Steven

. 1999. Foundations of English grammar. 5th edn. Leuven & Amersfoort: Acco.

12.

Depraetere

Ilse

Cappelle

Bert

Grandin

Cyril

Leclercq

Benoît

. 2021. Necessity modals and the role of source as a predictive factor. Paper presented at ISLE 6, University of Eastern Finland, Joensuu.

13.

Depraetere

Ilse

Reed

Susan

. 2011. Towards a more explicit taxonomy of root possibility in English. English Language and Linguistics 15(1). 1-29.

14.

Depraetere

Ilse

Reed

Susan

. 2021. Mood and modality in English. In Aarts

Bas

Hinrichs

Lars

McMahon

April

(eds.), The handbook of English linguistics, 207-229. 2nd edn. Oxford: Blackwell.

15.

Depraetere

Ilse

Verhulst

. 2008. Source of the modality: A reassessment. English Language and Linguistics 12(1). 1-25.

16.

De Smet

Hendrik

Verstraete

Jean-Christophe

. 2006. Coming to terms with subjectivity. Cognitive Linguistics 17(3). 365-392.

17.

Furmaniak

Grégory

. 2020. On the (con)textual properties of must, have to and shall: An integrative account. In Pascal

Hohaus

Schulze

Rainer

(ed.), Re-assessing modalising modal expressions: Categories, co-text and context, 281-310. Amsterdam: John Benjamins.

18.

Huddleston

Rodney

Pullum

Geoffrey

. 2002. The Cambridge grammar of the English Language. Cambridge: Cambridge University Press.

19.

Kastrone

Inga

. 2008. Need to and the modality of obligation: A corpus-based analysis. Oslo: University of Oslo Master’s thesis.

20.

Kratzer

Angelika

. 1981. The notional category of modality. In Eikmeyer

Hans-Jürgen

Rieser

Hannes

(eds.), Words, worlds, and contexts: New approaches in word semantics, 38-74. Berlin: Mouton de Gruyter.

21.

Kratzer

Angelika

. 2010. Modals and conditionals. Oxford: Oxford University Press.

22.

Larreya

Paul

. 1982. Quelques remarques sur have to et must. In Lavédrine

Jean

(éd.) Travaux XXXV, 103-121.Saint-Etienne: CIEREC.

23.

Larreya

Paul

Rivière

Claude

. 2019. Grammaire explicative de l’anglais. 5th edn. Paris: Longman Pearson.

24.

Leech

Geoffrey

. 2003. Modality on the move: The English modal auxiliaries 1961-1992. In Facchinetti

Roberta

Krug

Manfred

Palmer

Frank

(eds.), Modality in contemporary English, 223-240. Berlin: Mouton de Gruyter.

25.

Leech

Geoffrey

. 2013. Where have all the modals gone? An essay on the declining frequency of core modal auxiliaries in recent standard English. In Marín-Arrese

Juana I.

Carretero

Marta

Arús Hita

Jorge

van der Auwera

Johan

(eds.), English modality: Core, periphery and evidentiality, 95-116. Berlin: Mouton de Gruyter.

26.

Leech

Geoffrey

Hundt

Marianne

Mair

Chris

Smith

Nicholas

. 2009. Change in contemporary English. Cambridge: Cambridge University Press.

27.

Lyons

John

, 1977. Semantics, vol. 2. Cambridge: Cambridge University Press.

28.

Mair

Christian

. 2015. Cross-variety diachronic drifts and ephemeral regional contrasts: An analysis of modality in the extended Brown family of corpora and what it can tell us about the New Englishes. In Collins

Peter

(ed.), Grammatical change in English world-wide, 119-146. Amsterdam: John Benjamins.

29.

Mair

Christian

Leech

Geoffrey

. 2021. Current changes in English syntax. In Aarts

Bas

Hinrichs

Lars

McMahon

April

(eds.), The handbook of English linguistics, 249-276. 2nd edn. Oxford: Blackwell.

30.

Meyer

David

Zeileis

Achim

Kurt

Hornik

. 2003. Visualizing independence using extended association plots. In Kurt

Hornik

Leisch

Friedrich

Zeileis

Achim

(eds.), Proceedings of the 3rd international workshop on distributed statistical computing. https://www.R-project.org/conferences/DSC-2003/Proceedings/

31.

Müller

Friederike

. 2008. From degrammaticalisation to regrammaticalisation? Current changes in the use of need. Arbeiten aus Anglistik und Amerikanistik 33(1). 71-94.

32.

Myhill

John

. 1997. Should and ought: The rise of individually oriented modality in American English. English Language and Linguistics 1(1). 3-23.

33.

Nokkonen

Soili

. 2006. The semantic variation of need to in four recent British English corpora. International Journal of Corpus Linguistics 11(1). 29-71.

34.

Nokkonen

Soili

. 2010. How many taxis there needs to be? The sociolinguistic variation of need to in spoken British English. Corpora 5(1). 45-74.

35.

Nuyts

Jan

. 2012. Notions of (inter)subjectivity. English Text Construction 5(1). 53-76.

36.

Palmer

Frank R

. 1990. Modality and the English modals. 2nd edn. New York: Longman.

37.

Quirk

Randolph

Greenbaum

Sidney

Leech

Geoffrey

Jan

Svartvik

. 1985. A comprehensive grammar of the English language. London: Longman.

38.

Schützler

Ole

Herzky

Jenny

. 2021. Modal verbs of strong obligation in Scottish Standard English. English Language and Linguistics 26(1). 1-27.

39.

Smith

Nicolas

. 2003. Changes in the modals and semi-modals of strong obligation and epistemic necessity in recent British English. In Facchinetti

Roberta

Krug

Manfred

Palmer

Frank

(eds.), Modality in contemporary English, 241-266. Berlin: Mouton de Gruyter.

40.

Sweetser

Eve

. 1990. From etymology to pragmatics: Metaphorical and cultural aspects of semantic structure. Cambridge: Cambridge University Press.

41.

Taeymans

Martine

. 2004. An investigation into the marginal modals dare and need in British Present-day English: A corpus-based approach. In Fischer

Olga

Norde

Muriel

Perridon

Harry

(eds.), Up and down the cline – The nature of grammaticalization, 97-114. Amsterdam: Benjamins.

42.

Tagliamonte

Sali A.

2004. Have to, gotta, must: Grammaticalization, variation and specialization in English deontic modality. In Lindquist

Hans

Mair

Christian

(eds.), Corpus research on grammaticalization in English, 37-55. Amsterdam: John Benjamins.

43.

Tagliamonte

Sali

Smith

Jennifer

. 2006. Layering, competition and a twist of fate: The deontic modality system in English dialects. Diachronica 23(2). 341-380.

44.

Verhulst

Depraetere

Ilse

Heyvaert

Liesbet

. 2013. Source and strength of modality: An empirical study of root should, ought to and be supposed to in Present-day British English. Journal of Pragmatics 55. 210-225.

45.

Verstraete

Jean-Christophe

. 2001. Subjective and objective modality: Interpersonal and ideational functions in the English modal auxiliary system. Journal of Pragmatics 33(10). 1505-1528.

46.

Westney

Paul

. 1995. Modals and periphrastics in English: An investigation into the semantic correspondence between certain English modal verbs and their periphrastic equivalents. Tübingen: Max Niemeyer Verlag.

47.

Zufferey

Sandrine

. 2020. Introduction to corpus linguistics. Hoboken: Wiley.