Navigation – Plan du site

The unsaid in J. Joyce’s “Araby”: A linguistic perspective

Marcella Bertuccelli Papi
p. 17-28


L'implicite est un terme qui recouvre divers types de phénomènes, la plupart identifiés dans la linguistique et la sémiotique. Dans Implicitness in Text and Discourse (Bertuccelli Papi, 2000), nous avons émis l'hypothèse que les significations implicites peuvent être regroupées en trois catégories principales : l'inexplicite, l'impliqué et le subplicite, et nous les avons différenciées en termes d'ancrage, de ressources et de dérivation ou de processus inferentiel.
Dans cette étude, nous allons montrer comment ces catégories peuvent être exploitées dans l'analyse linguistique d'un texte littéraire, à savoir "Araby", tiré de Dubliners de James Joyce.
Nous mettrons en évidence comment des significations implicites se combinent pour créer une strate de non-dit, ce que nous appelons l'implicitude du texte. Celle-ci contribue à la profondeur sémantique et symbolique dont dépend la valeur littéraire de l'histoire.
La phrase apparemment innocente "O, I never said such a thing!", qu'entend le protagoniste de l'histoire pendant une conversation fortuite, peut être interprétée comme le point culminant à partir duquel se dévoile l'ensemble des propositions qui fonde la signification profonde de l'histoire.

Entrées d’index

Studied authors :

James Joyce

Notes de la rédaction

Traduit par Jean-Thomas Rieux

Texte intégral

1When linguistics encounters literature, the major risk is lack of mutual understanding. Linguistics is defined as the scientific study of language, where “scientific” implies, among other things, that it uses a metalanguage which is made for describing and explaining its object of investigation in an objective manner. In turn, literature claims by its very nature the right to infringe the rules and logic of ordinary verbal interaction, thus placing itself above and beyond the scope of “scientific” investigation.

2As a consequence, the impression of an unbridgeable gap arises and two extreme attitudes may result – namely, reductionism and/or mutual indifference. Mutual indifference is based on the assumption that what is useful for ordinary language is useless for literature. Reductionist approaches, instead, presuppose a view of literature as one of the various forms of communication, and, as such, subject to the same type of rules and constraints as ordinary language, albeit creatively exploited.

3The study of implicitness points to a third way of understanding the relationship between linguistics and literature, one which, while not negating the obvious fact that language is the shared core of the two domains, still assumes that if the power of literary expression is such as to allow a potentially unlimited number of interpretations, then there must be something in the underlying meaning which is never made totally and definitely explicit. Language is actualised in literary texts, but literary texts turn out to be complex inferential mechanisms (Eco 1979) pointing back to the nonfinite nature of language meanings. It is one of the aims of this paper to give substance to this claim.

4Implicitness is a complex, multifaceted and pervasive phenomenon whose understanding requires fine analytical tools and a very flexible theoretical apparatus. In linguistics, implicitness has been dealt with under various headings: entailments, presuppositions, implicatures, and more recently explicatures, and implicitures, are the most noteworthy.

5In Bertuccelli Papi 2000, I tried to show that the richness of the implicit cannot be exhausted in terms of these categories, and I argued that once we start to investigate where the line between the explicit and the implicit should be drawn, we are faced with phenomena that make it doubtful whether such a line actually exists. In fact, there is a shadowy area between the said and the unsaid where many phenomena are hidden, and subtler analytical categories are needed in order to deal with them. Consequently, I suggested that there is a continuum of implicit meanings ranging from what is said up to what is implicated, passing through various degrees and manners of saying and not saying.

6More particularly, I put forward the thesis that “implicitness” should be considered as a cover term for a wide range of phenomena, which I have grouped into three major categories, namely the inexplicit, the implicated, and the subplicit. These differ in terms of objects, resources and processes, and give access to different areas of textual interpretation. Since it would be impossible to summarize even rapidly the arguments that I discussed in the book in detail, I will only give a synthetic overview which I hope will be useful for understanding the goals outlined in the present paper.

7Essentially, the inexplicit concerns what is missing in order to have a full representation of a state of affairs or a situation; thus, ellipsis, semantic underdetermination and ambiguities typically belong to this category; knowledge of language and the co-text are the necessary resources for recovery of an explicit representation, and the two basic processes involved are fleshing out and filling in.

8 The implicated includes both the representation that putatively corresponds to the speaker’s meaning and what should be supplied as an implicit premise in an inferential reasoning leading to the representation itself; the distinctive character of an implicated implicit meaning is that it is formally unrelated to the explicit source from which it derives: if someone says “I have some work to do tonight” meaning “I cannot take the boys to the cinema”, the latter should be considered implicated as much as the implicit premise that had he been free, he would have taken the boys to the cinema. The resources are the multifarious forms of shared knowledge, and they are subject to selection via the principle of Relevance as defined by Sperber and Wilson 1987.  

9Finally, the subplicit involves all those shades of the unsaid that depend not so much on the content or the wording of the message as on perception of the attitude with which the message is produced. If a wife and mother of the child tells her husband “Your son is playing with his friend again”, she conveys not only the explicit information carried by her words, but also the subplicit meanings “I disapprove of it”, and possibly also “I dislike you” as a result of the strategy of detachment reflected in the choice of the referential expression “Your son” instead of “our son” combined with various premises from background shared knowledge.

10It is a necessary complement of my hypothesis that the three categories mentioned above (inexplicit, implicated and subplicit) include some core phenomena but have fuzzy boundaries - which accounts for the intuition that implicitness is in fact a continuum from surface to deeper configurations of meanings of growing complexity. This hypothesis seems to me to be strengthened by the literary text I have chosen to analyze, namely, “Araby” by James Joyce.


12The eleventh in order of composition of the stories that were to become Dubliners, “Araby” was explicitly conceived as the last in the first group, devoted to childhood-adolescence. That accounts for the numerous links with the two stories preceding it – “The Sisters”, and “An Encounter” – starting from the fact that the protagonist and narrator is a boy about the age of the protagonists in the other two tales, and further extending to the areas of religion, death, escape, relations with adults, meanness of everyday existence, up to the powerful unifying theme of paralysis.

13Intertextuality is certainly able to unveil a kernel of partly overlapping meanings which each story highlights from its own perspective, in a chiaroscuro interplay of themes from whose contrasts further meanings emerge. It would therefore be worth investigating the modalities in which microstructural implicit meanings contribute to the construction of macrostructural ones, but here I will mainly concentrate on a text-internal perspective.

14The story is familiar: Dublin in winter. A boy who lives with his aunt and uncle falls in love with a girl, the sister of one of his friends, living on the same floor. He has been in love with her secretly for a long time until finally, one day, the girls speaks to him and asks him whether he is going to Araby the following Saturday. The boy replies that if he goes he will bring her some little present. Since then, he yearns to see his dream come true, but unfortunately, because of some complications, he reaches the market too late at night, when the stalls are closing, and he sees himself “as a creature driven and derided by vanity”, his eyes burning “with anguish and anger”.

15The plot follows the canonical structure of the short story – exposition, rising action, climax, falling action, resolution /anticlimax – and it is rich in meanings of universal significance, interspersed with symbols that reinforce each of the sections in which it is articulated. But our interest here is not so much on the story which is narrated, as on the story that is left unsaid, which is hidden in the halo of implicit meanings that I have defined elsewhere as the “implicitude”of the text. Where does this story lie? What are the keys to its interpretations? There are numerous points in the narrative in which we become aware of the unsaid, and it is my task as a linguist to provide the instruments for their identification and the categories for their explanations.

16The first point I would like to make is a confutatio. Logico-semantic categories are not sufficiently fine-grained to serve as analytical tools for literary texts ’interpretation. Let’s consider two of the categories that are commonly cited in linguistic studies of the implicit – namely, entailment and presupposition.

1. Entailment

17The notion of “entailment" is a relation between semantic units, that is propositions, and it is drawn from classical logic. As such it is defined in terms of valid inferences, or, alternatively, in terms of truth values:

A entails B (A >> B) if B is true whenever A is true (or, in all worlds where A is true, B is true).

18This is a fundamental category in logic, because upon it all other logical relations, such as equivalence and contradiction, can be defined. Thus, the proposition underlying the sentence “Molly is a cat” entails the proposition underlying the sentence “Molly is an animal”: the former cannot be denied without also denying the latter


19Now consider the term “fib” which occurs in the final section of “Araby”:

‘O, I never said such a thing!’‘O, but you did!’‘O, but I didn’t’‘Didn’t she say that?’‘Yes, I heard her’‘O, there’s a fib!’

20Fib logically entails “not conforming to the truth”. Unfortunately, however, fib belongs to a set of near-synonyms all of which entail that the statement does not conform to the truth. The set includes lie, falsehood, untruth, misrepresentation: their core denotation is the same, but they differ from one another in fine aspects of their meanings. A lie is a deliberate attempt to deceive, a flat contradiction of the truth; a misrepresentation may be more indirect, a misplacement of emphasis, an untruth might be told out of ignorance, and a fib is deliberate but rather trivial, possibly told to save one’s own or another’s face. Moreover, there are stylistic differences among them: fib is an informal, childish term, whereas falsehood is quite formal, and untruth can be used euphemistically to avoid part of the derogatory implications of some of the other terms.

21Based on the identification of the logically necessary core of expressions, the notion of entailment is not subtle enough to account for the shades of meaning this small word introduces via the paradigmatic contrasts listed above. And this is a serious drawback, since the use of fib in this apparently petty dialogue carries with it such an amount of implicit meaning that no formal logic treatment would ever be able to explain. Fib is, to me, the key to the game that Joyce is playing with the reader precisely by writing things which are only partially true, innocent lies that hide deeper meanings that the reader can only detect by recognizing and interpreting such signs as this one (it is not by chance that these are precisely the words used by the boy when his uncle comes back home “I could interpret these signs”….).

22Still, entailments have long been discussed in linguistics and their role in establishing the internal consistency of texts is undeniable. Moreover, entailments have become particularly attractive in linguistics ever since the notion of presupposition started to be debated. Semantic theories, in fact, saw the possibility of formally defining presuppositions as a special type of entailment.

2. Semanchaud, supposition

23The notion of presupposition has been defined as an inference relation holding between statements, based on semantic entailment (or necessitation), and formulated as follows:

  1. A >> B (“A necessitates B” or “A semantically entails B”) if and only if whenever A is true, B is also true.

  2. A presupposes B if and only if both A and its negation entail B.

24According to this definition, semantic presuppositions came to be seen as a subtype of entailment – specifically, that type of entailment which remains valid even when A is negated. Therefore, given a sentence like

Papyrus stopped publishing poetry.

25we can infer that

Papyrus used to publish poetry.

26To prove that the inference is in fact a semantic presupposition, we apply the negation test, and see whether the inference can still be drawn. In fact

Papyrus has not stopped publishing poetry.

27still presupposes

Papyrus used to publish poetry.

28According to the definition provided above, “Papyrus used to publish poetry” is a semantic presupposition.

29The example shows that semantic presuppositions are assumed to be explicitly anchored to some linguistic form or expression of the utterance. In our case the anchorage is provided by the lexical item “stop”, but there are numerous expressions that can be reputed responsible for presuppositional inferences. These are called presupposition triggers.

30Presupposition triggers fall into two main categories: lexical and syntactic. They include definite descriptions, “factive” predicates, “implicative” verbs, change of state, inchoative and iterative verbs, clefting and pseudoclefting, temporal clauses, non restrictive relative clauses, counterfactuals. I will concentrate on definite descriptions and discuss their role as implicitness markers.

2.1. Definite descriptions.

31All definite descriptions, including proper names, are said to carry with them existential presupposition. Thus

Sue is watching the girl in the corner.

32presupposes that, in some possible world, there is a person called Sue, that in some place there is a corner,  and asserts that  there is a girl in that corner.

33Let us consider the proper names in our story. They include Araby, North Richmond Street, Christian Brothers’ School, The Abbot, Walter Scott, The Devout Communicant, The Memoirs of Vidocq, Mangan, O’Donovan, Mrs. Mercer, The Arab’s Farewell to his Steed. A philosophical tradition holds them to be “rigid designators”: they do not mean, they designate, and semantically presuppose the existence of their referents in some universe of discourse, in our case the fictional world created by the story. Unfortunately, however, this analysis is again a gross oversimplification of the enourmous amount of meaning implicitly conveyed by the names in “Araby”.

34For one thing, definiteness cannot be equated with existential presuppositionality tout court: at least, we should distinguish between existence and reference. If some entity is mentioned in a text, we may remain at the surface level, and assume its referent to be information which the writer poses as part of the knowledge given by the text, and as such existing within that universe of discourse. Alternatively, we may want to go deeper, and start to delve into our encyclopaedic knowledge, thus passing from an existential to a referential presupposition, that is to say trying to “identify” the referent within some other context of knowledge, say our “cultural” knowledge. Even further, we may wonder what its relevance is in the text, thus attributing an intentionality to the writer. Finally, we may connect such text-internal intentionality to some higher-order intentionality from which further implications may arise.

35It is clear that what I have in mind is a cline of implicit meanings only a fraction of which is covered by semantic presuppositions. Let me explicate in greater detail.

36It makes a difference in depth of understanding whether we simply assume by semantic presupposition that North Richmond Street exists in the story as the street where the boy lived, or whether we have access to the additional information that when Joyce was about the age of the boy in the story, he lived at 17 North Richmond Street, and that there was a “Grand Oriental Fete” in Dublin that ran from May 14 to 19, and that the theme song of this had such lines as

“I’ll sing thee songs of Araby,
And takes of fair Cashmere,
Wild tales to cheat thee of a sign,
Or charm thee to a tear…”

37And it takes us even farther to know that Joyce attended the Christian Brothers’ School for some time, rather than simply assuming the textual existence of the school.

38Even the choice of the book titles must go beyond the mere semantic presupposition that they really existed. In fact, all the books have real referents, but depending on the amount of cultural-encyclopaedic knowledge activated, different levels of access to implicit meanings will be accessed, and in fact they are accessed only once we ask ourselves what the relevance of such knowledge is to some theme of the story.

39The Abbot, written in 1820, was about Mary Queen of Scots, usually portrayed as a “harlot queen” and here described in a religious and romantic tone (the confusion of romantic, religious and physical love which pervades “Araby” is hereby hinted at); three works have the title The Devout Communicant, but possibly more important than identifying which one Joyce had in mind is the fact that any of them may have contributed to the choice of the boy’s devoutly pious vocabulary and outlook. The Memoirs of Vidocq, published in 1829, was again a real novel, dealing with a dishonest Parisian Police Commissioner who was also a thief. There is a meaning in the mention of this book, too, of which the (learned) reader should be aware, while the boy is not, and from this contrast further implications arise, clustering around one of the crucial themes of the story, namely that of deception.

40Whatever the actual implicit associations, we all feel that the choice of Mangan’s sister as the only identifier for the beloved girl is not by chance: Joyce could certainly count on his readers making the connection with the popular, sentimental romantic 19th century Irish poet, James Clarence Mangan (1803-1849), who was himself fond of writing about Araby and claimed that some of his poems were translations from Arabic.

41And if Jeremiah O’Donovan (1831-1915) really existed – a revolutionary who advocated the use of violence against British rule – what would be the purpose of the semantic presupposition that a Mrs. Mercer existed, while we perceive that the signifiant is so rich in connections with the imagery and theme of the mercantile and the mercenary in the story, and the explicit, fictional information that she is the widow of a pawnbroker further supports the implicit meanings triggered by the name?

42Finally, The Arab’s Farewell to his Steed actually has a referent in a poem by Caroline Norton (1808-1877), which tells of an Arab boy who sells the thing he loves most in the world, his horse, in exchange for a few gold coins, but can we ignore that Caroline Norton was herself the object of “commerce” in that her husband was said to have “sold her” to the British Home Secretary to Ireland by his silent complicity in an arrangement for his own professional gain? Not if we wish to unveil the deepest semantic recesses of the story’s soul. But this type of information comes to the surface once we have made a decision that it is somehow relevant to the thematic coherence of the story, which shifts the focus from semantic to pragmatic presuppositions or, as I prefer to think, from the area of the inexplicit to the areas of the implicated and the subplicit. This is clearly not to say that we should abandon the notion of presupposition tout court – rather, we should be careful not to overextend the term "presupposition" to all phenomena that are even vaguely related to what is not said explicitly, because in so doing we would make it coextensive with the process of interpretation itself, at least in the sense that interpreting is always going beyond the letter of what is said.

43It is on the basis of such considerations that I proposed a different view of implicitness. In order to account for the “implicitude” of the text, it seems to me more useful to distinguish areas gradually merging into each other, the discriminating points being marked by foci of relevance in the global coherence of the text. And since I believe the literary complexity of a text is enhanced when its global coherence is the result of multiple underlying coherences, and I regard coherence as crucially hinging upon implicitness, it follows that the overall complexity of a text is a function of the implicit meanings shaping both the local microsystems of coherence and the overall structure of the text.

44Let me present two examples, the first one from the microsystem of personal reference, the second from the microsystem of time coordinates. If we look at the referential expressions in “Araby, we find that the main characters – the boy and the girl – are never identified by their proper names: they are mainly referred to by personal pronouns: I/me/myself, we/us, he/himshe/her. As all definite descriptions, personal pronouns typically presuppose the existence of their referents, albeit identifiable text-internally, given the underdetermined semantic nature of personal pronouns (they are only able to identify a male referent and a female referent). The absence of proper names marks a degree of inexplicitness whose symbolic relevance emerges from the referential underdeterminedness of the pronouns. From the point of view of a referential semantics, however, it does not make any difference whether they have a subject or direct/indirect object function. In contrast, the choice is pragmatically meaningful, in that they crucially contribute to the development of the implicit story: the boy is initially immersed in the plurality of a “we” and gradually acquires the traits of his own subjectivity by turning into an agentive “I” in the relation with the girl, who is in turn initially presented as a patient-object “her”. It is only in the climax “At last she spoke to me” that the girl turns into a “she”, thus becoming an active subject in the story. From that moment onward, the boy is an “I”, an acting, thinking, loving, feeling “I”, until he reaches Araby, and here something important happens, which is again recorded by the pronouns: another “I” is introduced – it is the young lady talking and laughing with the two young men. The boy continues to be a subject “I”, but is no longer an agentive “I”, he becomes a semantic experiencer, the subject of stative verbs, until the final reflexive form “myself” comes to portray him as a patient of his disappointment and deception: “I saw myself as a creature driven and derided by vanity: and my eyes burned with anguish and anger”.

45Finally, the inexplicit-implicated dynamics may be observed within the domain of temporal coherence. If we look at the time indicators, we notice a sort of “telescoped” narration, starting with background information about “the short days of winter”, and “every morning”, then coming to a foregrounded “one evening”, and reaching the climax with “at last she spoke to me”. No one really wonders what else the boy did before the girl spoke to him, beyond what we are told by the writer. But after the encounter, we get the impression that we know what happens in the boy’s mind every single moment - time seems to slow down, the focus is on a single day, which is mentioned as a Saturday, specified as “Saturday morning”, and further segmented into short intervals - “after eight o’clock”, “at nine o’clock”, “ten past ten”. This scansion of times, while leaving inexplicit what happens in the “meanwhile”, iconically suggests the boy’s anxiety as the achievement of his goal gradually approaches. As a consequence, we perceive the unsaid as filled with moments of tension which contribute to the overall impression of deep emotional involvement only by remaining unsaid.

46It goes without saying that this is only a fragment of the analysis that could be carried out on implicitness in Joyce’s “Araby. I have not touched upon the problems of semantic ambiguity and nonliterality, nor have I tackled the problem of subplicit meanings as connected with the overall attitude of the writer, here represented by the range of symbols finely investigated by literary critics. I hope, however, I have given insights of the specificity of the linguistic perspective and, at the same time, of the large number of suggestions that literary texts are able to give to linguistic investigation.


Bertuccelli Papi, M. Implicitness in text and discorse, Pisa, ETS, 2000.

Eco, Umberto Lector in fabula, Milano, Bompiani, 1979.

Sperber, D.-D.Wilson Relevance, Oxford, Blackwell, 1987.

Pour citer cet article

Référence papier

Marcella Bertuccelli Papi, « The unsaid in J. Joyce’s “Araby”: A linguistic perspective », Journal of the Short Story in English, 40 | 2003, 17-28.

Référence électronique

Marcella Bertuccelli Papi, « The unsaid in J. Joyce’s “Araby”: A linguistic perspective », Journal of the Short Story in English [En ligne], 40 | Spring 2003, mis en ligne le 29 juillet 2008, consulté le 18 juin 2018. URL :


Marcella Bertuccelli Papi

Marcella Bertuccelli Papi is Professor of English Linguistics at Pisa University. Her current research interests include semantic and pragmatic aspects of meaning, text and discourse structures, the relationship between language and cognition. She is the author of several papers on a wide variety of subjects, extending from the history of language to syntax, semantics, lexical analysis and theoretical linguistics. She is a member of the scientific committee of the European Linguistics Society and of the editorial board of the “Journal of historical pragmatics”. She has recently published a monograph on Implicitness in text and discourse.

Droits d’auteur

© All rights reserved

  • OpenEdition Journals