Navigation – Plan du site

Surely not!
Between certainty and disbelief1

Graham Ranger

Résumés

Le marqueur anglais surely possède au moins trois fonctions différentes : adverbe de manière intraprédicatif, adverbe épistémique extraprédicatif et adverbe de discours. Dans cet article je proposerai une caractérisation unique de surely dans le cadre de la Théorie des Opérations Énonciatives (cf. Culioli, 1990, 1999a et 1999b). Nous verrons que, selon des paramètres contextuels variables, surely prend l’une de ses trois valeurs possibles. Plus précisément, surely marque une correspondance entre, d’un côté, une trajectoire préconstruite menant d’un point de départ vers un point d’arrivée (de (p, non-p) vers p), et, de l’autre, la même trajectoire construite dans la situation d’énonciation par l’énonciateur. Lorsque la trajectoire porte sur le mode de réalisation d’un procès, surely prend sa valeur intraprédicative ; lorsque la trajectoire porte sur le passage entre une situation repère et une situation projetée, surely prend sa valeur extraprédicative, comme adverbe épistémique. Le fonctionnement en tant qu’adverbe de discours de surely pose quelques problèmes pour l’analyse. Lorsqu’il est épistémique, le marqueur surely semble signaler une certitude, mais lorsqu’il est adverbe de discours, il semble plutôt signaler le doute, l’incrédulité ou l’incompréhension, selon le contexte. Je propose que lorsque surely s’emploie discursivement, il marque la prise en charge d’une trajectoire préconstruite par l’énonciateur, tout en reconnaissant l’existence d’une trajectoire préconstruite contre-orientée, prise en charge par un autre énonciateur (éventuellement le coénonciateur). Les glissements entre certitude et doute peuvent s’expliquer comme la conséquence d’une prise en compte d’un contexte discursif élargi. Les différentes valeurs contextuelles de l’adverbe de discours surely (cf. Downing, 2001) découlent, entre autres, de la façon dont l’énonciateur situe son discours par rapport à d’autres instances énonciatives.

Haut de page

Texte intégral

1. Introduction: what makes surely problematic

  • 1 My thanks for their encouragements and suggestions to the two reviewers of Discours.
  • 2 The term “marker” is a theoretical choice, based on the idea that linguistic it (...)

1The marker2 surely poses a number of interesting problems for the linguist, touching on questions of transcategoriality, grammaticalization theory, aspect, modality and argumentation. In Present Day English surely possesses at least three distinct and clearly differentiable values, illustrated below.

[1] Gazza is slowly but surely learning the Italian language, making it work for him.
  (BNC CH3)

2Here, surely functions as an intrapredicative adverb, contributing to determine the manner in which the process learn the Italian language is realised.

[2] Unless the Soviet military intervenes, self-determination must surely lead to reunification.
  (BNC A87)

3In example [2] above, surely is an epistemic adverb, working extrapredicatively to qualify the chances of realisation of the predication in question, i.e. self-determination / lead to reunification. We will later see that in such cases it can be hard to distinguish between epistemic values and purely intensifying values.

[3] O’REILLY: Now Ms. McLean, as a Democrat, I mean obviously all the Democratic presidential candidates are against this, but surely you understand the point that if American lives are endangered, most Americans I believe would say you got to waterboard them. You got to protect American lives.
  (COCA Fox O’Reilly Spoken)

4Lastly, in example [3] clause-initial surely is employed as a discourse marker by which the speaker both indicates commitment to a given proposition while at the same time situating this relative to opposing opinions that a co-speaker – real or potential – might hold. Insofar as this function goes beyond the boundaries of the clause, often implying a confrontational or argumentative context, surely here corresponds to Schiffrin’s classic definition of a discourse marker as “sequentially dependent elements which bracket units of talk” (1987: 31) or, more precisely, to the definition proposed by Fraser (1988: 21-22) for whom: “discourse markers […] signal a comment specifying the type of sequential discourse relationship that holds between the current utterance – the utterance of which the discourse marker is a part – and the prior discourse”.

5Each of these uses of surely has affinities with positions within the clause. Intrapredicative surely is invariably found in collocations with other manner adverbs in expressions of the slowly but surely kind. Epistemic surely typically occurs in medial position. The discourse marker surely is – in common with many discourse adverbs of its type – often placed clause-initially.

  • 3 The Oxford Advanced Learner’s Dictionary includes a category apart for surely “used with (...)

6Interestingly, while the epistemic adverb surely tends to reinforce the degree of speaker commitment to a given proposition, as a discourse marker, surely can often give the impression that, although the speaker would very much like to believe something, the evidence to the contrary rather undermines his or her confidence. This is particularly noticeable when surely is in negative contexts, of the surely not type illustrated in [4]3:

[4] She stared at him in silence, chilled by the frozen wastes reflected in his eyes, and he said raspingly, “The fires of the damned, wasn’t that how you put it? Is this what you wanted? Does this satisfy your thirst for revenge?” Lissa swallowed hard, remembering her words. Surely he did not blame her for this?
  (BNC HA6)
  • 4 The Oxford Advanced Learner’s Dictionary or the Longman Dictionary of Contemporary English, for (...)

7In a thorough and insightful study of the discourse marker function of surely, Downing (2001) cites Biber and Finegan (1988) or Halliday (1985) who, in common with the practice of many dictionaries4, relate surely to adverbs like certainly. As Downing remarks, however, among the category Biber and Finegan unhappily term “the surely adverbials”, “surely is in a sense the odd one out” (2001: 253). Indeed, if the epistemic surely of [2] might admit reformulation with certainly, the same is not true of the discourse marker surely of [3], or [4]:

  • 5 The conventions used in manipulations are as follows: → indicates a manipulatio (...)
[2a] → […] self-determination must certainly lead to reunification5.
[3a] → ≠ […] but certainly you understand the point that if American lives are endangered, most Americans I believe would say you got to waterboard them.
[4a] → ≠ Certainly he did not blame her for this?

8Unlike certainly, surely cannot be used, in standard British English in any case, in short answers:

  • 6 In spoken examples from the BNC involving more than one locutor, I have added q (...)
[5] “Do you want to tell us all about the stall for the energy conservation bill?” / “Certainly. Yeah the stall was extremely successful.”
  (BNC JP7 Spoken6)
[5a] → ?? “Surely. Yeah the stall was extremely successful.”

9Even more surprisingly, perhaps, for an adverb which on the face of it appears to reinforce the speaker’s commitment to a proposition, surely – unlike certainly – very frequently occurs in interrogative contexts:

[6] They leave the skin on? You don’t have to eat the skin, surely?
  (BNC KBW Spoken)
[6a] → ?? You don’t have to eat the skin, certainly?
[7] “Would you like to come down to Carinish Court for a few days?” “You wouldn’t want me there.” Peter was aware of his heart thudding in his chest. “Not at Christmas, surely?”
  (BNC CKB)
[7a] → ?? Not at Christmas, certainly?
  • 7 Aijmer (2002), in a comparative study, makes a similar point: “Surely as a disc (...)

10Again in the words of Downing: “Surely is not what it seems; it dresses up speaker’s opinion in the form of a query or an exclamation, for despite the fact that it occurs in statements, it elicits a reply, as do tags” (2001: 253)7.

11To sum up the ground covered so far:

12Firstly we noted the potential polyvalence of surely, which can function as an intrapredicative adverb, as an extrapredicative, epistemic adverb and as a discourse marker.

13Secondly, we saw that, despite its classification together with certainly and other, related expressions of certainty, surely possesses a number of semantic and syntactic features which clearly set it apart from would-be synonyms.

14The aim of the following paper is to show that the different values of surely can be explained if we postulate that surely marks a single, constitutively underdetermined operation which can be parametered in different ways, enabling us to calculate and predict the various values it assumes in discourse.

15In section 2 I will lay down one or two theoretical preliminaries before proposing the complex enuncative operation of which surely is the marker. In section 3 I apply this to a selection of authentic examples, by way of illustration. Our examples will be drawn essentially from the British National Corpus and the Corpus of Contemporary American English.

2. A schematic form for surely

16The explanations that follow are formulated within the framework of the Theory of Enunciative Operations elaborated by Antoine Culioli and his collaborators over the past forty years or so. Initial work with the theory focused on single-speaker, clause-internal issues, to simplify, but recently the theory has been applied with some success to questions of discourse and discourse markers. It is, in any case, in the nature of the theory not to remain limited to one predetermined area of language activity but to seek out diverse aspects of linguistic communication in an integrative approach.

17In the Theory of Enunciative Operations the term marker is used, not just for discourse markers, but as a general term for linguistic forms (morphemes, constructions, prosodic features…) seen as markers of operations. The point is that a linguistic form is not conceived of as mapping directly onto a meaning or series of meanings, but as marking a more abstract, invariable operation, which is shaped by contextual parameters to provide a specific value. The goal of the linguist is to endeavour to reconstruct the invariance postulated behind the diversity of values, as well as to be able to account for the process by which an invariant operation – known as a schematic form – allows us to construct a range of values – or shapes – in context. As Culioli (1990: 178) writes:

My contention is that utterances display shapes that derive from complex forms which the linguist reconstructs through modelling. The goal is to lay bare the formal workings that underlie the production and the recognition of interpretable shapes, namely, utterances. We have no access to the processes that originate the forms on which the shapes are based, but we have, at our disposal the textual traces which point to such processes.

18In the case in point, our goal is to reconstruct for the marker surely the abstract schematic form associated with the various shapes in which surely appears. We will consider these shapes to be those of the intrapredicative manner adverb surely, the extrapredicative epistemic-intensive adverb surely and the discourse marker surely. Within each subtype, there will be room for further variation, but this threefold division provides an expedient way of organising our demonstration.

19Two key concepts of the Theory of Enunciative Operations will prove particularly useful to us in the course of the presentation: the concept of the branching path, and of the enunciative source.

2.1. The branching path

20Central to linguistic activity is the operation of categorisation, as speakers seek to match prelinguistic, cognitive representations (or, what a subject means) and linguistic representations (how to say it). This is presented within the Theory of Enunciative Operations as an operation whereby occurrences are situated relative to a complex topological space called a notional domain.

  • 8 The notation non-p is linguistic, not logical and so can mean p is not the (...)
  • 9 The source for this type of representation are the seminars of Culioli, translated into (...)

21A notional domain comprises, at the very least, an Interior (I), corresponding to an operation of identification with the notion, p, and an Exterior (E), which we may note non-p, corresponding to an operation of differentiation relative to the notion8. It is also possible to construct a Boundary area, where occurrences may possess properties of p and of non-p, and a Centre, which provides a regulating occurrence for p, either in terms of a typical occurrence (the organizing centre) or in terms of an ideal, maximal occurrence (the attracting centre). This may be represented graphically as follows9:

22The positions so far described all allow us to situate occurrences somewhere on a notional domain. We can also construct an off-line position noted IE or (p, non-p), on a different plane altogether, from which both the Interior and the Exterior are potentially accessible via opposing trajectories. This is useful when dealing with, for example, interrogatives, or questions of modality, including surely, in the present case. The branching path model is represented graphically below:

2.2. The enunciative source

23The Theory of Enunciative Operations differentiates between the enunciative source (the instance of endorsement of the utterance, which we will call the speaker), the locutor (which we might, for the sake of argument, call the talker) and the grammatical subject. By default, the speaker and the locutor are the same, i.e. the person talking – the locutor – also assumes responsibility for what is said, but the two may be different, when reading out loud, in reported speech, for ironic purposes, in gnomic utterances and so on. A speech situation implies a speaker, a co-speaker (whether present or absent) and a time (and place) of utterance. Conventionally these concepts are noted S0 (speaker), S0’ (co-speaker), Sit0 (speech situation) and T0 (time of utterance). Additionally, we will be referring to a fictive speaker S01, constructed independently of S0 and S0’, which can be an absent speaker or a generalised enunciative instance – any speaker, or the speech community.

2.3. Hypothesis for an operational description of surely

24We consider that surely marks an operation of identification between, on the one hand, a preconstructed trajectory leading from a start point to an end point, and on the other, the same trajectory constructed in the speech situation by the speaker.

25This is, in the terms of the previous section, the schematic form which, it is hoped, will enable us to explain the various values of surely in context as configurations of a fundamental invariant operation depending on, in this instance, the type of trajectory involved and the way in which the speaker positions his or her discourse relative to that of the co-speaker and the speech community.

26I will be looking, firstly, at the intrapredicative manner adverb (3.1) before moving on to the epistemic adverb (3.2) and the discourse marker (3.3). Lastly, I will consider the etymological origins of surely in the light of several less common uses (3.4).

3. Case studies

3.1. Intrapredicative surely

27Intrapredicative surely was illustrated above by example [1]:

[1] Gazza is slowly but surely learning the Italian language, making it work for him.
  (BNC CH3)

28The vast majority of the examples of intrapredicative surely in Present Day English involve the collocation slowly but surely, or related expressions. We might end things there, considering this as a sort of fossilised idiom, possibly favoured by the alliterative qualities of the collocation. However, I think it is important to reflect upon how the combination slowly but surely contributes to determine the predication.

29I posit that in such utterances surely marks a qualitative conformity with a preexisting representation despite a potential non-conformity, resulting from the quantitative determination with slowly. In [1], for example, which is fairly typical of the type, the process learn a language involves a start point and an end point, corresponding to the beginning and the end of the learning activity, and defining a trajectory. This can be represented in the following manner:

30The first marker, slowly, implies reference to a model of realisation, a rate at which the process learning a language might typically take place and which is not met in the current circumstances. In other words, slowly means more slowly than one might expect, more slowly than normal. This quantitative non-conformity relative to a preexisting typical representation might lead us to infer that the process is in some way qualitatively imperfect, i.e. that the normal trajectory leading continuously from start point to end point is in some way not followed as it should be. The determination with surely identifies the effective trajectory with the anticipated trajectory, thereby reestablishing qualitative conformity with the model.

31Correspondingly, slowly but surely type idioms generally occur with processes of a telic nature, in other words, processes which are notionally bounded, as further illustrated by the examples below:

[8] All the time the train is carrying us slowly but surely up the French coast.
  (BNC F9R)
[9] Slowly but surely the rights of trade union reps are being eroded.
  (BNC HLU)
[10] The Italian Federation is fighting a rearguard battle against this trend, which is slowly but surely killing Italian rugby.
  (BNC CB2)

32In each case the processes carry up the coast, be eroded or kill Italian rugby involve a preexisting representation of a passage from start point to end point of the type illustrated above. The marker surely identifies the effective trajectory with the preexisting representation, despite the potential for differentiation contained in slowly. The next example involves the more unusual collocation swiftly and surely:

[11] Still another legacy of Vietnam is the conviction that a war must be fought swiftly and surely. “If we go in, we have to win”, said Steven Schmidt of Salt Lake City, the owner of several camera and bicycle stores, who served in the Air Force in Vietnam.
  (COCA 1991 NEWS New York Times)

33The argument presented for slowly and surely can nonetheless be maintained here: excessive haste, like excessive slowness, can also threaten the qualitative conformity with a preexisting model and so, as before, surely marks an operation whereby a trajectory, corresponding to the path from the start point of the process fight a war to the end point, is identified with a preconstructed trajectory. Interestingly, the collocation with swiftly, though widely attested in the COCA, is absent from the BNC, suggesting a separate development of intrapredicative surely in US and British English, which will be confirmed in the case of epistemic surely, too.

34In positional terms, idioms of the slowly but surely type appear free to occur clause-initially [9], clause-medially [1] and [10] and clause-finally. Generally speaking, one would expect a manner adverb to occur most often clause-finally. In slowly but surely collocations, however, surely is more than a simple manner adverb. In fact the whole group functions as a quasi-concessive, within which surely contributes to reinforcing the predication, despite the potential opposition vehicled by the associated adverb. Even in collocation with manner adverbs, then, surely already assumes certain modal functions.

3.2. Epistemic surely

  • 10 The modals typically involved are must, can and will as one would expect.

35When surely constructs epistemic values, it is invariably in medial position, where it collocates with certain modals, as in [12] and [13]10:

[12] Unless the Soviet military intervenes, self-determination must surely lead to reunification.
  (BNC A87)
[13] But those committees need to be really careful to make sure if they don’t want a big storm to be stirred up here, that any of the embryos that are used clearly have been placed beyond the pale of being fertilized before their use. There are a large number of embryos that we know are never going to be fertilized, where the people who are in control of them have made that clear. The research ought to be confined to those. And I think the committees will surely do that.
  (COCA CNN Newsroom Spoken)

36The medial position of surely is predictable since this is where one expects to find adverbs which determine the realisation of the predication. Additionally, surely often forms a single tone-group with the accompanying modal, which it contributes to determine. In such examples surely often appears to reinforce the speaker’s commitment to the projected predicative relation without any reference to other, potentially counter-oriented, perspectives. This is particularly clear if we try to move surely into clause-initial position:

[12a] Surely self-determination must lead to reunification.

37If the position of surely within the proposition was irrelevant, then [12] and [12a] would be equivalent. This is not the case, since [12a] is clearly more appropriate in a polemical context, in the presence of possible or anticipated contradiction.

38The epistemic comment clause I think in [13] associates easily with epistemic surely but appears to be incompatible with surely as a discourse marker, as the following manipulations show:

[13a] ?? And I think surely the committees will do that.
[13b] * Surely I think the committes will do that.

39Whichever clause we decide to place surely in front of, the result is unsatisfactory.

40If we agree, then, that epistemic surely constructs different values from the discourse marker surely, then we might compare [12] to [12b], in which surely has been deleted:

[12b] Self-determination must lead to reunification.
  • 11 Cf. Deschamps (1999) for this type of representation applied to the system of modal aux (...)

41How can we differentiate surely must from must alone? Following the branching path representation of modality we can consider that must indicates a necessary path from (p, non-p) to p, and excludes – or bars – the path from (p, non-p) to non-p11.

42The segment must surely indicates a supplementary operation whereby the speaker identifies the trajectory determined in the speech situation with surely must to the trajectory preconstructed with must. In doing so, much as was the case with slowly but surely, the speaker indicates that the passage from a start point ((p, non-p) in this case) to an end point (p) corresponds to a preexisting model.

43The difference with intrapredicative surely is that, whereas with intrapredicative surely we were dealing essentially with an aspectual path leading from the beginning of an event to its end, here, we are dealing with the modal trajectory from the speech situation to a projected event situation. Similar arguments apply to will surely [13], can surely for example, which also construct trajectories from (p, non-p) to p.

44It is worth noting that [12b] self-determination must lead to reunification can be brought closer to [12] self-determination must surely lead to reunification if the modal is accentuated, typically with a high-fall, i.e. self-determination ↓must lead to reunification. This is normal if we consider that, like surely, stress of this type is also used to mark reidentification.

45The value surely constructs here is a reinforcing value, such that the probability appears even greater with must surely than with must alone. This reinforcing value does not however apply well to the next example:

[14] “[…] Do you have any music you could play?” “I surely do.” Bill rooted amongst a rack of laser discs.
  (BNC HTU)

46Here I surely do might be reformulated as Indeed I do / You bet etc. This type of use is considerably more prevalent in American English than in British English. These remarks might lead us to wonder how a similar configuration might produce, in some utterances, a reinforcing effect (almost certainly, very probably), in others, an intensive effect (indeed).

47I would suggest that the values vary according to the nature of the domain qualified by surely. If the trajectory from the speech situation to the projected event situation is constructed as gradable, then surely assumes reinforcing values (really) while if it is constructed as non-gradable, i.e. an all-or-nothing choice, then surely constructs an intensive effect. This type of variability is common enough, as, for example, in the different uses of quite, according to the type of adjective qualified: quite/fairly interesting compared with quite/absolutely fascinating. In our example, the modals must or will allow for degrees of probability, i.e. gradability, while do constructs a bipolar choice, hence, in surely do utterances, epistemic surely typically assumes intensive values, reinforcing speaker commitment.

48I mentioned the fact that such a usage appears characteristic of American English. In [15] – one of only four examples of the sequence I surely do in the BNC compared with 35 in the COCA – surely is used in this way, in a caricature of American speech (by an English author, incidentally).

[15] “Hi, there”, he said, with his widest grin. She looked up and smiled back. Tall, stooping, tweed hat, Burberry and calfskin grip –; an all-American tourist. “Good afternoon, sir. Can I help you?” “Well, now, I hope so, miss. Yes, I surely do. Ya see, I just flew in from the States and I took your British Airways –; my all-time favourite airline –; and you know what they did? They lost my luggage. Yes, ma’am, sent it all the way to Frankfurt by mistake […]”.
  (BNC CAM)

49And in [16], what is meant reassuringly in American English sounds rather less so in British English!

[16] She whispered in his ear, “Do you really love me, Cable?”
“I surely do, Mrs. Jordan”, he answered with an easy grin.
  (COCA Fiction)
  • 12 Downing, 2001: 260. More generally, one might criticise the fact that Downi (...)
  • 13 Interestingly, Downing’s BNC example of this use: “And you’ve come all the way from Ame (...)

50Downing recognises this category: “A minor use, shared by surely and sure expresses a strong affirmation, often as an affirmative response equal to ‘yes’”12. Although I agree with the link she makes between this use of surely and adverbial sure, I do not agree that this intensive use of surely is particularly minor; it is just not characteristic of British English13.

51To sum up, in this section we have seen that, as with its intrapredicative uses, the epistemic adverb surely still marks an identification between a trajectory constructed in the speech situation and a preconstructed trajectory. The difference is that, in intrapredicative uses, the preconstructed trajectory is essentially aspectual, leading from the beginning to the end, whereas in extrapredicative epistemic uses, the preconstructed trajectory is modal, leading from the speech situation to a projected event situation. Within this category, American English allows a further, intensive use of surely, which derives from the use of surely to quantify a non-gradable, bi-polar domain.

3.3. The discourse marker surely

52For the discourse marker surely we will of course retain the schematic form presented in 2. The difference, in the case of the discourse marker, is that the trajectory involved here is the one the speaker chooses when he or she endorses a propositional content for a given situation. This perhaps requires some explanation: a speaker who asserts a propositional content p, makes the choice to assert p in preference to non-p. This choice can be represented in a similar way to modality, with a trajectory passing from a preassertive level (where a choice is available), to an assertive level (where a choice has been made) and subjective endorsement of a propositional content.

53The discourse marker surely indicates that the speaker maintains their preconstructed endorsement of the proposition, thereby identifying their enunciative choice in passing from (p, non‑p) to p to a previous choice of endorsement in a movement which Downing revealingly refers to as “speaker self-validation” (2001: 256). When used as a discourse marker, surely is typically in clause-initial position, but may be medial or final. Wherever it is placed, however, it forms an independent tone-group and receives prosodic focus.

54Let us look at an example with this model in mind:

[17] Owen was keen to show me his new electric racing car set and let me by the arm to his room. I was taken aback by the decor –; vivid red, yellow and green geometric patterns screaming from every wall. Surely a most inappropriate colour scheme for the bedroom of a hyperactive child?
  (BNC B06)

55One might – independently of the context of [17] – imagine that an excessively vivid colour scheme is inappropriate for a hyperactive child. In saying Surely a most inappropriate colour scheme […], the speaker reaffirms his or her adhesion to this preconstructed proposition p. Now, a reaffirmation is generally made in the face of potential opposition, here the position non-p, presumably endorsed by Owen’s parents. At the same time, the speaker attempts to enlist the support of an as yet undecided co-speaker, who is encouraged to take the same path from (p, non-p) to p. The three speakers involved in the configuration here correspond to the speaker, the co-speaker and a third person, a potential speaker, absent from the speech situation. We might represent this graphically as follows:

56This use of surely is described in Downing’s paper as a case of the speaker inviting “affirmation or corroboration from an addressee regarding the state of mind, intentions or actions of a third party” (2001: 268). In fact, when it is used as a discourse marker surely allows the speaker to construct a wide range of argumentative values “wishful thinking”, “challenge” or “self-questioning”, to name but three of the categories mentioned in Downing’s paper. My hypothesis is that these values depend crucially on how the speaker S0 positions his or her endorsement of p relative to the co-speaker S0’ and an absent, or fictive, speaker S01. The next examples provide further illustration.

[18] “I already know you as well as I could ever want to. And besides, it would take a stronger stomach than mine to contemplate furthering our acquaintance.” As he simply smiled at her reaction, she hurried on to ask a question of her own. “But surely you’re not intending to move in here permanently? You already have a house in Edinburgh.”
  (BNC JXA)

57Here, we have a construction of the surely not type, included among Downing’s “challenge” category. These typically involve, in addition to the identification of the (p, non-p) → p trajectory, associated with S0, a preconstructed position non-p attributed to the co-speaker S0’, which we could represent as follows:

58In other words, the speaker reaffirms endorsement of p in the face of opposition from the co-speaker, to whom the speaker attributes the counter-oriented proposition non-p.

[4] “We had a visitation, a welcoming note from a pyromaniac.” She stared at him in silence, chilled by the frozen wastes reflected in his eyes, and he said raspingly, “The fires of the damned, wasn’t that how you put it? Is this what you wanted? Does this satisfy your thirst for revenge?” Lissa swallowed hard, remembering her words. Surely he did not blame her for this?
  (BNC HA6)

59This illustrates a similar type of opposition between a co-speaker positioned in non-p, and a speaker, who reaffirms the choice leading from (p, non-p) to p. Interestingly, if the following context makes it clear that this is free indirect speech, such that the speaker Lissa is addressing herself to the unnamed male co-speaker, then we have the same challenging configuration as before, if this is a passage of free indirect thought, however, then we are closer to the previous example, as Lissa is both speaker and co-speaker, seeking to convince herself of something. This enunciative split, as a speaker questions her own judgement, produces concomitant implications of self-doubt, another of Downing’s categories of meaning.

[19] O’REILLY: Now Ms McLean, as a Democrat, I mean obviously all the Democratic presidential candidates are against this, but surely you understand the point that if American lives are endangered, most Americans I believe would say you got to waterboard them. You got to protect American lives.
  (COCA Fox O’Reilly)

60In this example, the speaker O’Reilly, in a live debate, knows that his co-locutor/co-speaker Ms McLean is opposed to the form of torture called waterboarding, but seeks nonetheless to enlist her endorsement by reenacting, as it were, his passage from (p, non-p) to p. He also presupposes that the speech community (most Americans) think the same way, in his use of understand: understand x implies the existence of an x to be understood. It is as if Ms McLean were one of the last opponents to p and the speaker O’Reilly the voice of the speech community S01. We might represent this additional configuration as follows:

  • 14 Note that in the theory of enunciative operations the subject pronoun I, for example, d (...)

61Downing (2001) pursues corpus searches, essentially using the subject pronoun criterion to classify different values, while admitting that the relationship between subject pronoun and the value of surely is not automatic. Although Downing’s intuition is important here, I think the essential determining factor is not the subject pronoun as such14, but the way the speaker positions his or her own discourse – and the re-endorsement of p in particular – relative to other, potential speakers: the co-speaker, a fictive speaker or the speech community. These positionings can be seen as further variables allowing us to derive specific values for the discourse marker surely from the complex interplay of a schematic form with elements of context.

  • 15 Downing writes: “there are three meaning of surely which I believe to be cr (...)

62The key difference between the approach adopted here and that of Downing is that, while Downing considers surely to carry a number of meanings in and of itself15, I prefer to derive these from the interaction of an abstract schema with contextual determinations. Let us illustrate this with a further example. One of the core meanings of surely, for Downing, is “mirativity”. This “refers to the linguistic marking of an utterance as conveying information which is new or unexpected to a speaker” (2001: 256). [20] below, which forms part of the title of Downing’s study, illustrates such a mirative use:

[20] “Surely you knew!”

63Unlike Downing however I would prefer to argue that this mirative effect derives from the properties of surely, associated with the process know in the second-person. The process know localises a propositional content (X) relative to a subject. It is a subjective verb insofar as only the grammatical subject is qualified to say whether he or she possesses the knowledge in question. In other words, one might have the firm conviction that one’s co-speaker knows something, but one cannot actually know that they know it. In utterances of the Surely you knew! type, the speaker appears to maintain a preconstructed endorsement of p (you knew X) even though the only speaker really qualified to affirm p is in fact the co-speaker you. This interplay between surely and its context leads naturally enough to a polarisation of enunciative positions and to values which can go from genuine surprise, or “mirativity”, to downright disbelief.

3.4. Remarks on the etymology of surely

64There exist one or two infrequently attested uses of intrapredicative surely without an associated and potentially counter-oriented adverb, and generally with markers of comparison, which can be linked diachronically to current uses, in the light of grammaticalization theory.

[21] The massive pin-sharp image completely fills the five-storey screen and you are there, as surely as if you were in an orbiting Shuttle.
  (B3K)
[22] Or would it be a sign of still greater maturity for their staff to go on contributing to a national system, a system in which the collaboration of the entire academic community could raise standards higher and judge quality more surely?
  (BNC HTK)
  • 16 The substitution of securely for surely is admittedly more convincing in [21] (...)
  • 17 Cf. for example, Traugott, 1995 or 1997.

65It is perhaps here that surely is closest to its etymological origins, Lat. securus, and it may indeed be paraphrased with securely16. In etymological terms, surely and sure both appear to date back to fourteenth-century Middle English, deriving from cognates of Modern French sûr and sûrement. Generally speaking, grammaticalization theory indicates a movement from more precise, lexical meanings to vaguer, procedural meanings17. Aijmer and Simon-Vandenbergen, for example, claim without discussion: “Like other adverbs, including certainly, surely has developed diachronically from a manner adverb” (2007: 135). Although this affirmation doubtless represents a general tendency, I am not certain it can apply in this form to surely, which was already remarkably polyvalent in Middle English, as the examples below show:

  • 18 These examples are taken from the online corpus at the University of Michigan: http://q (...)
[23] Wythouten doubte trew obedyencers folow surely oure lorde & his wordes where he seyth.
  (From Three Middle-English Versions of the Rule of St. Benet and Two Contemporary Rituals for the Ordination of Nuns18)
  • 19 The translation provided is my own, and is meant simply as a gloss, to indi (...)
[23a] → true followers follow our lord and the words he says unfalteringly19
[24] […] þis medicyne shalle make yow hoole surely, as men seyn.
  (John Russells, Boke of Nurture, Harl. MS. 4011, Fol. 171)
[24a] → this medicine will definitely make you better, as people say
[25] “Have ye seen how the kynge durste not ete all this souper / for fere that mawgys shold werke wytchecraft vpon hym” / “Surely”, sayd rowlande, “it is true”.
  (From The Right Plesaunt and Goodly Historie of the Foure Sonnes of Aymon. English from the French by William Caxton, and printed by him about 1489)
[25a] Indeed, said Rowland, it is true.
[26] “Cosyn, what folke is yonder, as ye thynke / for it semeth a grete oost a fore Iherusalem / are thei sarrasyns or crysten? what saye you?” “Surely”, sayd mawgys, “I canne not telle, and I am sore merveylled what it maye be”.
  (Ibid.)
[26a] To tell the truth, said Mawgis, I cannot tell, and I have no idea what they might be.

66In fact, the development of surely is relatively complex. When sure/surely entered English, French sûrement had already developed a number of discursive uses. Additionally, Old and Middle English possessed the cognate term sickerly (O.E. sicerlice), deriving ultimately from Lat. securus via West Germanic, which had developed discursive uses of its own. Finally, securely enters sixteenth-century English, supplanting the intrapredicative manner adverb surely in the majority of cases, perhaps making it easier for surely to function modally. Hence, in English at least, it appears that surely has never functioned uniquely as a manner adverb but has always, under the probable influence of Fr. sûrement and M.E. sickerly, always possessed a range of epistemic or discursive uses.

  • 20 From the Oxford English Dictionary entry for the adjective and adverb sure.

67It is easy to conjecture on the links between the etymological senses of “Free from or not exposed to danger or risk; not liable to be injured or destroyed”20 and the schematic form proposed in section 2. In each case what is involved is the elimination of potential differentiation between a preexisting representation that the speaker judges to be “right” and a given specific representation.

4. Conclusion

68In the preceding pages I hope to have shown that surely functions in a variety of different ways in contemporary English, as an intrapredicative adverb (generally in collocations of the slowly but surely type), as an epistemic adverb (clause-medially and in association with a modal) and as a discourse marker.

69I propose to attribute a single, invariant operation to surely, a schematic form, from which we can derive specific values in context. I consider that surely marks an operation of identification between, on the one hand, a preconstructed trajectory leading from a start point to an end point, i.e. from (p, non-p) to p, and on the other, the same trajectory constructed in the speech situation by the speaker.

70When the trajectory concerns:

  • the movement from the start point to the end point of a process, then surely is intrapredicative;
  • the movement from the speech situation to the projected event situation, then surely is epistemic;
  • the subjective endorsement of a proposition, then surely is a discourse marker.

71When surely functions as a discourse marker it can vehicle a wide range of speech acts, from aggressive challenge, to persuasion, to expressions of self-doubt or surprise. These, I have suggested, can be calculated as different configurations of the schematic form, depending upon how the speaker positions his reaffirmed endorsement of a proposition relative to the co-speaker, to other absent speakers or to the speech community.

Haut de page

Bibliographie

Aijmer, K. 2002. Modal Adverbs of Certainty and Uncertainty in an English-Swedish Perspective. Language and Computers 39: 97-112.

Aijmer, K. & Simon-Vandenbergen, A.-M. 2007. The Semantic Field of Modal Certainty: A Corpus-Based Study of English Adverbs. The Hague: Mouton de Gruyter.

Biber, D. & Finegan, E. 1988. Adverbial Stance Types in English. Discourse Processes 11: 1-34.

Culioli, A. 1990. Representation, Referential Processes and Regulation. In Pour une linguistique de l’énonciation. Paris – Gap: Ophrys. T. 1: 177-213.

Culioli, A. 1995. Cognition and Representation in Linguistic Theory, M. Liddle (ed.). Amsterdam – Philadelphia: John Benjamins.

Culioli, A. 1999a. Pour une linguistique de l'énonciation. Paris – Gap: Ophrys. Vol. 2: Formalisation et opérations de repérage.

Culioli, A. 1999b. Pour une linguistique de l'énonciation. Paris – Gap: Ophrys. Vol. 3: Domaine notionnel.

Davies, M. 2004-. BYU-BNC: The British National Corpus. Available online: http://corpus.byu.edu/bnc.

Davies, M. 2008-. The Corpus of Contemporary American English (COCA): 410+ Million Words, 1990-Present. Available online: http://www.americancorpus.org.

Deschamps, A. 1999. Les modaux de l’anglais. In A. Deschamps & J. Guillemin-Flescher (eds), Les opérations de détermination : quantification / qualification. HDL. Gap: Ophrys.

Downing, A. 2001. “Surely You Knew!”: Surely as a Marker of Evidentiality and Stance. Functions of Language 8 (2): 253-285.

Fraser, B. 1988. Types of English Discourse Markers. Acta Linguistica Hungarica 38: 19-33.

Halliday, M.A.K. 1985. An Introduction to Functional Grammar. London: Edward Arnold.

Hornby, A.S. 2005. Oxford Advanced Learner’s Dictionary. Oxford: Oxford University Press [7th edition].

Schiffrin, D. 1987. Discourse Markers. Studies in Interactional Sociolinguistics 5. Cambridge: Cambridge University Press.

Summers, D. (dir.) 2003. Longman Dictionary of Contemporary English. Harlow: Longman [4th edition].

Traugott, E.C. 1995. Subjectification in Grammaticalization. In S. Wright & D. Stein (eds), Subjectivity and Subjectivisation. Cambridge: Cambridge University Press: 31-54.

Traugott, E.C. 1997. The Role of the Development of Discourse Markers in a Theory of Grammaticalization. Paper presented at ICHL XII, Manchester 1995. Available online: http://www.stanford.edu/~traugott/ect-papersonline.html.

Haut de page

Notes

1 My thanks for their encouragements and suggestions to the two reviewers of Discours.

2 The term “marker” is a theoretical choice, based on the idea that linguistic items are the traces (or markers) of cognitive operations to which the linguist has no direct access.

3 The Oxford Advanced Learner’s Dictionary includes a category apart for surely “used with a negative to show that something surprises you and you do not want to believe it” (Hornby, 2005: 1544).

4 The Oxford Advanced Learner’s Dictionary or the Longman Dictionary of Contemporary English, for example.

5 The conventions used in manipulations are as follows: → indicates a manipulation; ≠ represents a change in meaning; ?? an utterance which poses problems of acceptability; * an unacceptable utterance.

6 In spoken examples from the BNC involving more than one locutor, I have added quotation marks for clarity.

7 Aijmer (2002), in a comparative study, makes a similar point: “Surely as a discourse marker is associated with uncertainty and questioning while certainly as a discourse marker indicates emphasis” (Aijmer, 2002: 109).

8 The notation non-p is linguistic, not logical and so can mean p is not the case, other-than-p is the case etc.

9 The source for this type of representation are the seminars of Culioli, translated into English in Culioli (1995).

10 The modals typically involved are must, can and will as one would expect.

11 Cf. Deschamps (1999) for this type of representation applied to the system of modal auxiliaries in English.

12 Downing, 2001: 260. More generally, one might criticise the fact that Downing excludes from her article consideration of any other uses of surely than as a discourse marker.

13 Interestingly, Downing’s BNC example of this use: “And you’ve come all the way from America to write about Heymouth?” “I surely have” is evidently a stereotypical representation of American speech.

14 Note that in the theory of enunciative operations the subject pronoun I, for example, does not refer directly to the speaker, but marks an operation identifying the subject of the utterance with the speaker.

15 Downing writes: “there are three meaning of surely which I believe to be crucial: surely as a marker of self-validation, surely expressing surprisal or mirative meaning, and surely as a way to foresee and so forearm oneself against foreseen denial” (2001: 257).

16 The substitution of securely for surely is admittedly more convincing in [21] than in [20].

17 Cf. for example, Traugott, 1995 or 1997.

18 These examples are taken from the online corpus at the University of Michigan: http://quod.lib.umich.edu.

19 The translation provided is my own, and is meant simply as a gloss, to indicate the way in which surely is being used in the Middle English context.

20 From the Oxford English Dictionary entry for the adjective and adverb sure.

Haut de page

Pour citer cet article

Référence électronique

Graham Ranger, « Surely not!
Between certainty and disbelief », Discours [En ligne], 8 | 2011, mis en ligne le 11 juillet 2011, consulté le 16 août 2017. URL : http://discours.revues.org/8416 ; DOI : 10.4000/discours.8416

Haut de page

Auteur

Graham Ranger

ICTT (Identité culturelle, textes et théatralité), EA 4277
Université d’Avignon et des Pays de Vaucluse

Haut de page