The Concept of Literature as a Problem for Literary History

As is well known, the word ‘literature’ can also carry non-literary meanings: ‘everything written’, ‘the writings on some specific subject’ and more. Yet what can be called ‘the literary concept of literature’ is commonly considered to be one and only one thing. The Oxford English Dictionary defines literature in the literary sense as ‘written work valued for superior or lasting artistic merit’ (Oxford English Dictionary 2024: literature, n., 3.b.). Other major anglophone dictionaries offer similar descriptions of the word’s meaning in literary contexts. And I do not doubt that the literary concept of literature is thought of as being one and only one by users of English, and of the other languages in which it is found.

Still, as with so many of our notions, when one starts to look closer, much ambiguity appears. With respect to the literary concept of literature, I am not thinking of the circumstance that different language users will entertain different ideas about what, more precisely, literature actually is, or that different specialists may offer their own various definitions. Instead, I want to point to the fact that there are different conventional, institutionalized, ways of taking the concept, different established practices for using it. According to major dictionaries, not only the OED, ‘literature’ is an evaluative concept: writings will have to be good in order to qualify as literature. But this is not how the notion is used in libraries, since a value-neutral concept of literature is obviously needed for the purposes of library classification.Footnote ^c Furthermore, the use of the concept in literary studies is often wide and vague, and those who award literary prizes apply their own criteria for eligibility.Footnote ^d

What we have here are entrenched differences in use that would, according to normal linguistic principles, be reflected in dictionary definitions if these were more ambitious than the present ones. The one-ness of the concept of literature is a reality in one sense, but at the same time an illusion. Works can be literature according to one established usage while being non-literature in some other.

One may wish to object that the important question ought to be how the concept of literature is employed by specialists within literary studies, and that the use of the concept in libraries, in connection with literary prizes, etc., must be a side issue. I do not agree, and, as I just observed, the academic use of ‘literature’ is often wide and vague.Footnote ^e Anyway, the one special split in the literary concept of literature on which I will focus is a split within academic literary studies itself, more precisely, within academic literary history.

In literary history writing, the concept of literature is traditionally used much more broadly about times before the late eighteenth and early nineteenth centuries than about more modern periods. For example, older works of history-writing, philosophy and oratory will conventionally be included in a literary history if they are important enough, while more modern history-writing, philosophy and oratory will not. With respect to ancient Greece, a world history of literature will conventionally have things to say about Plato’s Symposion, Thucydides’ history of the Peloponnesian War, and the orator Demosthenes, while, when it comes to twentieth-century Britain, it will bypass the philosophical writings of Bertrand Russell, and also Winston Churchill’s speeches and his history of the Second World War (see, for example, von See Reference von See1972–Reference von See2008 (index in vol. 25) and Damrosch and Lindberg-Wada Reference Damrosch and Lindberg-Wada2022 (index in vol. 4)).

With regard to more ancient civilizations than that of classical Greece, the inclusiveness of the concept of literature will be even greater. As Miriam Lichtheim (Reference Lichtheim1973: vi) writes in the introduction to her anthology of ancient Egyptian literature:

One of Lichtheim’s specimens of ancient Egyptian literature is a royal decree by a pharaoh inscribed on a stela (Lichtheim Reference Lichtheim1973: 28).

This all means that a literary history that also covers times before the eighteenth century, and particularly a world history of literature, will apply different criteria for inclusion depending on the time and place under consideration. Where there is interesting poetry, drama or fictional prose to account for, these texts will certainly be brought in, but there will be considerable variability where other material is concerned. I can see no rational motivation for this (but I will return to it later). This feature of the literary history genre will rather have a historical cause.

It is a fact that, in Roman culture, poetry, oratory, history and philosophy were the most important prestige genres (Fantham Reference Fantham1996: 17; but cf. Kenney Reference Kenney, Kenney and Clausen1982: 5–6). These were the kinds of writing of which all truly educated people were expected to have knowledge and experience. The early writers of literary history will have been quite familiar with Latin and with the Roman way of thinking about verbal compositions. They will have found it natural to include not only the poetry of Virgil and Horace, but also Cicero’s speeches and his discourses on philosophy and rhetoric, and Caesar’s historical writings, and to make analogous choices regarding India and China and ancient Egypt. While this practice cemented itself in the writing of the literary history of older times, the new, narrower concept of literature became successively more established during the nineteenth century and more and more self-evident as the concept to use about more recent texts. The two ways of employing the literary concept of literature were thus drawn apart, and the use of two different logics became an accepted trait of the genre of literary history.

This explanation of how the temporal split in the literary concept of literature came about is, admittedly, a hypothesis. I have not undertaken the concrete research that would be needed in order to turn the account just given into a proven truth. But my main point in this section is that there is, conventionally, a difference between the wider sense in which the concept of literature is used about older times and the narrower sense employed about later periods. This much is obvious and undeniable. In the rest of this article, I will reflect on how this discrepancy is to be assessed and what recommendations for the writing of literary history one can justifiably give with respect to this problem.

The Idea that Literature Exists and is the Given Object of Literary History

One might think that there is an obvious way of overcoming the discrepancy to which I have been referring. After all, literature exists. The natural strategy could seem to be to find out what literature really is and then write the history of that, and only that, which is truly literature. But things are not so simple. In my view, this is not because it is difficult to find out what literature really is, but because there is nothing that literature really is, not in the sense presupposed by the strategy just sketched.

‘Exist’ is another tricky word. When philosophers theorize about what ultimately exists, what the world ultimately contains, the kind of existence they are thinking of is, typically at least, so-called mind-independent existence.Footnote ^f What exists mind-independently is that which exists no matter what people think or say: that which might be there even if humankind were to disappear. The usual prime example of what exists mind-independently is the physical world – not the physical world such as it is described by us, but the physical world such as it is in itself – but some philosophers would include abstract entities such as, for instance, the natural numbers.

Be that as it may. The point I want to make is that what we think of as existing are often things that do not possess mind-independent existence but thoroughly rely on human thought, often also on human agreement, for their very being. The contents of our daydreams exist, in this mind-dependent sense, but we can change these contents at will. To take something much more consequential and complicated as an example: the European Union exists, but its existence is wholly dependent on an intricate pattern of changeable, and changing, human agreements, so the EU, too, exists by human fiat. If humankind ceased to be, so would the European Union. Mind-independent things, such as the physical stuff that we refer to as the Justus Lipsius building in Brussels, might still be there, but not the mind-dependent human institution that we call the European Union.

Literature exists, but the division between literature and non-literature – or, rather, the many criss-crossing divisions alluded to in the previous section – must be mind-dependent, human-made. There cannot very well be a division between literature and non-literature that forms part of brute physical (or abstract) reality itself. I will not belabour this point by going deeper into philosophical ontology, but will take it, in this context, as something self-evident. I leave it to those who do not share my view to explain how a mind-independent division between literature and non-literature could have come about and persisted as part of the structure of the universe.

My excursion into philosophy can seem uncalled-for, but understanding something about the complexities of the notion of existence is in fact important for the literary historian. It is all too easy to take our current descriptions of the world around us not just as our accepted way of thinking and speaking, but as reflecting the true properties of the brutally existing world. We use the concept of literature, we speak of compositions from many times and cultures as literary, and we speak of fewer genres as literary when we consider more modern times. It is easy to think that this is not just the language we use but the true way of seeing things. But if the distinction between literature and non-literature is mind-dependent, it is something we can, in principle, change at will. There will be no question of what literature is in any absolute sense, but rather a question of how it seems reasonable – on such-and-such practical or theoretical grounds – to draw the boundaries around it.

Histories of literature mostly ignore the discrepancy between older and newer literature to which I keep referring, at least in their general presentations of aims and methods. The Russian Istorija vsemirnoj literatury (A History of World Literature, 1983–1994) is an exception. The difference in the scope of literature in different parts of the work is in fact commented on in one of its introductory articles, written by Nikolaj I. Konrad, a prominent orientalist and one of the main theorists behind the Istorija. However, for Konrad, what I call a discrepancy has to do with changes in literature itself. For him, literature simply exists, it has a history, and during the course of its history it has come to comprise fewer genres than before. Konrad views this as a very significant fact about the history of literature: he writes that ‘the historical changes in what is included in literature is one of the most important phenomena in its history’ (Konrad Reference Konrad and Braginskij1983: 15, my translation). Where I see the result of a difference in our use of the concept of literature, he sees something truly existing that changes over time. If he were right, the problem that occupies me in this article would prove illusory and go away. But, as I just explained, I do not think that convictions such as Konrad’s can be upheld.

The Concept of Literature and Expectations on Histories of Literature

It can seem reasonable to use one, consistently applied concept of literature throughout a literary history. But, if so, what should that concept be like?

In his and Austin Warren’s Theory of Literature (1949 and later), René Wellek considered literature to be, first and foremost, poetry, drama and fictional prose, and he maintained that it is fictionality that makes these types of writing form a special category. This made him want to exclude nonfiction from literature. ‘If we recognize “fictionality”, “invention”, or “imagination” as the distinguishing trait of literature’, he wrote, ‘we think thus of literature in terms of Homer, Dante, Shakespeare, Balzac, Keats rather than of Cicero or Montaigne, Bossuet, or Emerson’ (Wellek and Warren Reference Wellek and Warren1966: 26). But thoughts of the demands of literary history caused him to later change his mind. In ‘What is literature?’ (Wellek Reference Wellek and Hernadi1978) he asked: ‘Shall one exclude Montaigne, Pascal, Burke, Gibbon, Berkeley, etc., from literature because they do not even pretend to write fiction?’ His answer was now negative: he suggested that ‘we, sensibly, include them in this concept (for how can we imagine a history of English literature in the eighteenth century without Gibbon, Berkeley and Burke?)’ (Wellek Reference Wellek and Hernadi1978: 21). Wellek did not take this line of thought further, so it is not clear whether or not he would want to use an older and wider concept of literature throughout a literary history – for example, whether or not he could imagine a history of English literature in the twentieth century that included Russell and Churchill. What is clear, though, is that including such writings would involve a break with extant expectations on the genre of literary history.

This would also be the case if one were to choose, instead, to apply a modern, narrower concept of literature throughout a literary history, as Wilt Idema and Lloyd Haft mean to do in their Guide to Chinese Literature (1997), a book that is, in practice, a history of Chinese literature. Pointing to the ‘wide variety of meanings’ that the word has today, the two authors emphasize the need for a kind of definition of ‘literature’ when the word has a structural role in a discourse: ‘if we wish to make serious use of the term “literature”, we must first make clear the sense in which we intend to use it’ (Idema and Haft Reference Idema and Haft1997: 3). I agree completely. For their own part, they opt for the modern, narrower concept. ‘In our study of Chinese literature’, they write, ‘we shall take the modern Western concept of literature as our guide’ (1997: 11). As they characterize it, this concept ‘views literature as a form of language use which primarily emphasizes the aesthetic function rather than the communicative function’ (1997: 3). This is clear enough, but it contravenes genre expectations, and Idema and Haft do not in fact hold on to the concept by which they claim to be guided. There is much about history and philosophy in their book, as one can see already by the fact that their Chapters 7 and 8, part of their account of the period before 100 ce, have the titles ‘Historical Prose’ and ‘The Philosophers’, respectively. Obviously, genre expectations have taken precedence over the authors’ stated aims here, imperceptibly it seems.

One might think that there must be a ‘right’, or at least a ‘best’, way of defining ‘literature’, and that it would be natural to employ this right or best concept in the writing of a literary history. Although it has often been said that ‘literature’ cannot be defined – something which is arguably both true and untrue, since it will depend on what you require from a definition – not so few qualified researchers have been searching for such a generally acceptable way of understanding the term. René Wellek was one of them, and there are several other good examples as well (see, for example, the contributions in Winko et al. Reference Winko, Jannidis and Lauer2009, and in Löck et al. Reference Löck, Urbich and Grimm2010, two comprehensive German collections of articles on the idea of literature). However, these attempts to define ‘literature’ do not normally attend to the multiplicity of the established ways of using the term and the many different purposes that they serve. Almost without exceptions, the one-ness of the literary concept of literature is too trustingly taken for granted. And, of course, even a ‘best’ definition of ‘literature’ (‘best’ for what purpose and according to what criteria?) would militate against traditional expectations on the genre if applied consistently in a history of literature.

Concluding Remarks

A discrepancy between the concepts of ‘literature’ used about older and more recent periods appears to be a constitutive feature of histories of literature as we know them. What conclusions for the genre are we to draw from this lack of logic in its very foundations?

One possible reaction is to just shrug off my remarks. If there really is a certain discrepancy, a certain lack of logic, so what? Literary histories have been written for well over 200 years and served us reasonably well, so why bother? Is not this much ado about nothing?

It is true that such things as world histories of literature, in their various present shapes, are valuable in very many ways just as they are. First and foremost, their inclusive and historically grounded overviews offer opportunities to consider the phenomena in which one is interested – be it an author or a group of authors, a certain motif, a type of work or the idea of literature itself – embedded in an abundance of historical material relevant for comparison and reflection. It is also true that all products of intellectual work have their limitations and inconsistencies. Still, when it comes to a history of literature, the concept of literature is not a detail too insignificant to pay attention to. If one sees and accepts that the concept is being used inconsistently in one’s history work, it could be argued that intellectual honesty demands that one should, at least, make this circumstance known to the reader and not just sweep it under the carpet.

Another possibility would be to employ a consistent concept of literature throughout. As the discussion of Wellek and of Idema and Haft above suggests, one might write the history of fictional discourse, or the history of verbal compositions meant for aesthetic appreciation. But it would most probably be viewed as misleading to present such overviews as histories of literature. Or one could, for example, define ‘literature’, quite nominally, as ‘poetry, drama and fictional prose’, and account for the history, across times and cultures, of that which one chooses to call poetry, drama and fictional prose. If one thinks that these macro-genres have something important that binds them together, such a history of literature would have its motivation in a wish for a certain definable kind of knowledge. This would be a history of literature in which neither pharaonic inscriptions nor older history, philosophy and oratory would be presented as literature, but they would not necessarily be absent. They could be there as a background to the focal, literary phenomena, which would not necessarily be lifted out of their cultural context but could be treated as parts of the wider textual world of their societies.

In any case, a truly important thing in any piece of scholarly or critical writing, a history of literature or not, must be to be clear, oneself, about what one is attempting to display or explain, about what picture of what realities one wishes to hold up to one’s readers. In my view, when planning a history of literature, perhaps a world history of literature, one should at least be attentive to, and adopt an attitude to, the intricacies in connection with the concept of literature to which I have pointed here.