Medieval literary genres are notoriously difficult to define, and medieval romance is especially so. After a century of scholarly effort has failed to produce a clear consensus, it is tempting to give up the game, to assume that everyone knows what a medieval romance is even if no one can define it, and to move on to the seemingly more attractive exercise of discussing the texts themselves. Kevin Whetter is to be commended for insisting that the issue needs to be revisited, that understanding genre is fundamental to understanding literature, and that generic conventions did form what Jauss famously called a "horizon of expectations" for authors and audiences of the medieval English romances--and this book is, despite the generality of its title, primarily concerned with Middle English romance.
The book has three main sections. The first defends the importance of understanding genre. Whetter points out, sensibly, that calling a romance a "mode" rather than a "genre" (after Northrop Frye) simply substitutes one vaguely defined term for another without resolving the issue. He acknowledges the difficulty of defining some genres: first, because definitions and conventions change as historical, cultural, linguistic, and literary contexts change--this is especially true of romance--but, second, and centrally to his argument, because the mixture of genres is very common. That genres change, or that a literary text may be a generic hybrid, is no reason to deny the reality or importance of literary genre as a contract between author and audience.
The second section of the book attempts to define medieval romance, primarily in contradistinction to the epic, and with a particular focus on Middle English verse romances. Here Whetter suggests that four characteristics are essential to the genre of romance: adventures are sought for their own sake rather than being uncomfortable necessities imposed by hostile forces; ladies play central rather than background roles; love becomes an important motivating factor; and all ends happily for the protagonist, who achieves material victory instead of perishing in a heroic last stand, however honourable. Few will disagree that adventure, ladies, and love are in some way characteristic of medieval romance, or that such familiar English texts as King Horn, Orfeo, Ywain and Gawain, or Launfal belong in this genre. More contentious is the insistence on the happy ending as an essential feature of romance, not merely as a common convention; not only does this requirement problematize the ending of Sir Gawain and the Green Knight, as Whetter notes (the cheerful knights of Arthur's court seem to assume that they are in a romance, but Gawain, guilt-ridden and scarred, is not so sure), but it also excludes from the genre all stories about the death of Arthur.
The third section of the book tests this definition of medieval romance against Malory's Morte Darthur. Maintaining, contra Vinaver, that Malory intended his Arthuriad as a "Hoole Book," Whetter demonstrates meticulously that Malory consistently turns his narrative toward tragedy--not only in the final part leading up to the disintegration of the Round Table and the death of Arthur, but throughout all his stories of Arthur and his knights, beginning from Arthur's own conception in a context of civil war, deceit, and illicit love. Because Malory so strenuously resists the possibility of a happy ending, Whetter suggests that this work is not a romance but a generic hybrid, a "tragic-romance" "with epic-heroic elements" (109). This is a detailed and careful reading of Malory, and the strongest part of the book.
The weakest part of the book, by contrast, is the middle section that defines medieval romance. Here I find problematic the strategy of defining romance against epic, not least because "medieval epic" is itself so difficult to define. Indeed, the opposition of romance and epic is largely a scholarly artifact, culturally and historically determined (in English scholarship it can be traced back to W. P. Ker's Epic and Romance of 1896), and associated with a questionable view of literary history in which the dignity of heroic epic was superseded by the frivolity of romance as the Middle Ages waned toward decadence. The medieval picture is less clear-cut. Epic was not a medieval English term as romance was; Middle English texts such as Cursor Mundi, Richard Coer de Lyon, and the Laud Troy Book list, among the subjects of romance, such "epic" figures as Alexander the Great and Charlemagne or events such as the siege of Troy. Furthermore, neither in its characterization of epic nor in its characterization of romance does this section of the book escape the suspicious circularity of extracting the essential features of each genre from texts chosen as exemplary precisely because they exhibit those features. A better strategy might have been to develop more fully a discussion of other types of evidence that Whetter does mention, but too briefly: manuscript layout, verse forms, and the contexts in which the word romance is used in Middle English.
The real problem is not the definition of romance presented here, but the book's assumptions about what the definition of a genre is supposed to do. As long as defining a genre means drawing a boundary around it so that we can decide which texts belong in the genre and which do not, we are doomed to frustration; but if defining a genre means describing its central tendencies, identifying its most influential texts, exploring its conventions, and elucidating its relationships with other genres--all without suggesting that there exists some master checklist of essential features that will determine absolutely whether membership should be conferred upon or denied to any given text--then we shall be in a better position to do what this book so convincingly represents as necessary: to understand literature by understanding genre.
Whetter's concept of hybrid genres recognizes, to some extent, the fuzziness of generic categories; and this account of medieval romance would be greatly enriched by more exploration of the relationships between romance and a wider range of other genres. Even in the discussion of romance and epic, there is not much advantage in contrasting English chivalric romance with Homer or Beowulf, when the romances coexisted with--and often were based on or derived from--French chansons de geste and Anglo-Norman poems about heroic founders of powerful dynasties. Other genres with which romance had a close relationship were allegory, saint's life, and history. This last relationship is particularly important to a consideration of Malory, for Caxton called his edition of Le Morte Darthur neither romance nor tragedy, but history. In doing so he was not committing himself unequivocably to the truth of Malory's narrative, but to its moral value, asserting that the stories there told were those worth remembering: "Historye is a perpetuel conservatryce of thoos thynges that have be doone before this presente tyme and also a cotydyan wytnesse of bienfayttes, of malefaytes, grete actes, and tryumphal vyctoryes of all maner peple," he wrote in the preface to his edition of the Polychronicon, further explaining that it is especially important to preserve a record of such events, so that virtue might be immortal even though men are mortal.  Although we cannot take Caxton's words as evidence of Malory's own intentions, they do suggest how a knowledgeable fifteenth-century reader would have regarded Malory's text. Romancers from Wace to Walter Scott have frequently considered the materials of history, real or imaginary, grist for their own narrative mills; and I would argue that Malory persistently deflects romance toward tragedy because he has committed himself to writing the whole history of Arthur, and, as medieval people knew, history is essentially tragic, because no one on this earth, not even a hero, lives happily ever after. I am not sure what we gain from creating a hybrid category of "tragic-romance" that cannot be achieved from acknowledging that the conventions of romance can be used, as Chaucer showed in Troilus and Criseyde, to set the scene for tragedy, or from admitting that romance might, after all, have an unhappy ending; the favourite subjects of medieval romance--Alexander, Troy, Charlemagne, Arthur--may allow small victories, but their master narratives end in defeat. 
Happily, Whetter's fine reading of Malory is not wholly dependent on the restrictive definition of romance that the book advocates; happily, too, my own dissatisfaction with the definitions put forward in the book has no bearing on the cogency of Whetter's argument that the definition--or at least the description--of medieval romance remains a critical necessity. This book will not be the last word on defining Middle English romance, let alone medieval romance more generally; but any brave attempt is to be valued for clarifying, to whatever extent, the horizons of our understanding.
1. N. F. Blake, ed., Caxton's Own Prose (London: André Deutsch, 1973), 130. The two "histories" that Caxton proposes to print, with this description in mind, are the Polychronicon and (!) the Golden Legend.
2. See especially Helen Cooper, The English Romance in Time: Transforming Motifs from Geoffrey of Monmouth to the Death of Shakespeare (Oxford: Oxford University Press, 2004), who devotes a chapter to "Unhappy Endings" (361-408).