To start with, one has to admit that Murray is to be commended for finding something new, and mostly true, to say on such well-explored topics as Chrétien de Troyes and translatio studii. Her book does not entirely carry conviction, however, despite the cleverness of most analyses; her readings of texts, both the almost too famous Erec et Enide and Chevalier de la Charette and relative rarities such as Piramus et Tisbé or Narcisse, are usually well-conducted and even brilliant, despite some translation mistakes, but the different parts do not really add to a satisfactory sum.
The departure of SJM's From Plato to Lancelot--provocation aside--is Plato's retelling of the Myth of Atlantis in his Timaeus. According to her, this retelling "gave learned twelfth-century authors an authoritative model for the creation of mythical--and yet, in some sense, true--poetic narratives." (xviii) It also provided these same authors with a warning and an incentive to write down stories, instead of just trusting in oral transmission. In choosing this Platonician perspective, SJM confers a new twist upon the somewhat harrowed-down motive of translatio studii, and the demonstration of the first chapter, "Reading Plato, Writing Romance," allies careful attention to the facts (knowledge of the manuscripts, composition of the libraries, etc) and intelligent interpretation. So does, for that matter, the second chapter in "Part 1: The Mediterranean," that deals with "Rewritings of Ovid," although one encounters the first contestable translations by SJM in the course of this original re-reading of two Ovidian poems. 
Unfortunately, the second Part, "The Atlantic," indispensable in order to build the balance that produces, according to SJM, the revolutionary conjointure created by Chrétien, is globally less convincing. Chapter 3, "Brendan's Voyage," rests largely on what one might call "criticism in the conditional form": SJM first formulates an interesting, albeit already slightly biased hypothesis: "Might not the Latin narration of Brendan's voyage, composed by an Irish monk who belonged to this 'Atlantic world,' be properly understood to a very significant degree as a response to Plato's Myth of Atlantis?" In the absence of incontrovertible written evidence (the Latin text in question mentioning Plato and the Timaeus), SJM has then to resort to further hypothetical bits, some of them probable, others less so. Phrases such as "they must have acquired" (98), "it is not difficult to imagine" (99), and numerous uses of "would" and "might" do not add up to a convincing argument: the hypothesis remains only an hypothesis.
Furthermore, the close readings of Benedeit's Voyage de Saint Brendan rest all too heavily upon the two rhetorical figures that SJM sets down to track in all her corpus: chiasmus and adnominatio. There are, undoubtedly, cases where these figures are used, knowingly or not, by twelfth-century authors. One may even suggest that Chrétien de Troyes relies a little too much on such devices to convey his intentions. However, chiasmus does occur fairly regularly in verse literature, and it may not be wise to be put too much emphasis on the meaning of such passages. As for adnominatio, such a figure is usually identified when at least five or six words belonging to the same paradigm occur in a short passage. To affirm, as SJM does once, that "These three occurrences constitute an adnominatio" (210), especially when these occurrences are spread on 41 lines and belong to the not-terribly-rare paradigm of description (descrivre, description), is somewhat naïve. Consequently, the frequent use of this critical tool--the identification of chiasmus and so-called adnominatio--does not enhance to the credibility of the argument.
To go back to Brendan: this chapter concludes with a few interesting remarks, but one does not feel that they logically follow on the previous analysis, the more so since this analysis is marred by some translation mistakes. The worst case in point is the affirmation that "At last, the pilgrims understand and embrace the divine truth (li veirs divins)" (125). Indeed, SJM has previously translated the line "Dunc dist Brandan li veirs divins:" with "And so Brendan said the divine truth" (line 918). Unfortunately, "li veirs divins" is not a cas régime, or an object in this sentence, but rather an epithet of Brendan himself, as shown by Merrilees' and Short's 2006 translation: "Brendan, en veritable homme de Dieu, déclara."  It is laudable to craft one's own translations when revisiting texts that may have become a little boring because of the very familiarity most readers have now with them. Substituting mistakes for perfectly valid translations is not so laudable.
Despite its poetic title ("The Wave Cry, the Wind Cry"), chapter 4 is perhaps the least convincing in the book. Following the meanders of what might be perceived as a kind of "reversed" translatio studii, via the return of Irish missionary saints and pilgrims to the Continent, SJM stresses "the influence [Merolilanus'] Celtic compatriots exerted in the development of a pan-European spiritual, intellectual, and cultural identity" (170). While this is undoubtedly true, the slightly disjointed line of narrative in this chapter does not quite manage to demonstrate it.
Maybe it is because, as interesting or convincing as they might be, these first two parts, and especially chapter 4, are mainly necessary to stage the scene for the real focus of the book: Part 3, "France," which concentrates on Chrétien de Troyes, or, more specifically, on Erec et Enide and to a smaller extent Cligs in chapter 5, "The Dawn of Arthurian Romance," and on Lancelot ou le Chevalier de la Charrette (and, a little, on Le Conte du Graal) in chapter 6, "Lancelot and the Future of Lettrere." These two chapters are mainly close readings of literary texts, and that is clearly what SJM does best. Her analyses are clever, precise and often original, not a small feat in presence of texts that have been studied so much. One may just regret the intrusion of comments somewhat out-of-place in an academic work (e.g. in one paraphrastic passage, after mentioning that Enide thinks Erec is dead, SJM adds: "Imagine her suffering!" (200) and, again, some mistranslations. 
In chapter 5, SJM takes over the acknowledged link between Martianus Capella's De Nuptiis Mercuri and Philologiae and Erec et Enide, and she succeeds in demonstrating how the structure of the latter is modeled on the other's. Of course, in view of her arguments, it is only logical that SJM justifies or interprets Erec et Enide's presence in Cligs' Prologue as a triumph of intertextual translatio. The only quibble I would have with this chapter has to do with the place given, or not given, to the Celtic homologue of Erec et Enide, Gereint.
Chapter 6, the final one, is logically enough the apotheosis of the book. Once again, it essentially consists of re-readings of a text so famous that one has almost forgotten how incredibly clever it is. SJM distances herself from the traditional interpretation according to which Chrétien, in the prologue of Lancelot, ostensibly demonstrates his resistance against the topic imposed by Marie de Champagne. The parallel between Chrétien/Marie and Lancelot/the queen is not only clever, but oddly convincing, even though the elements of Tristanian criticism do not bring much to the demonstration.
However, all of this seems strangely useless, in the light of both the conclusion of the chapter and the general conclusion that follows, having broadened the scope of the study to encompass Le Conte du Graal: "Marie de Champagne's France could not be more different than Plato's Greece, and yet Le Chevalier de la Charrette bridges these worlds and draws its strength from both of them." (252) Then again: "(Chrétien's) singular genius lay in his ability to conjoin both traditions [the Atlantic and the Mediterranean] in the creation of a new genre of thoroughly French vernacular story telling." (260) This is quite true, but was it necessary to go through all this to come to such unoriginal conclusions? All in all, however, SJM has brilliantly played her part in the pageant that develops around Chrétien de Troyes since 1185. Her study does not appear quite as necessary as earlier attempts may have been, but it is certainly clever, and it does bring some new grain to grind to the mill of twelfth-century romance studies.
I tend to find academic bibliographies, with their numerous subdivisions corresponding to different aspects of the topic studied in the book, somewhat confusing and cumbersome. In that case, however, I would have been in favor of partitioning the bibliography into two sections, one devoted to classical and medieval works, the other to "secondary literature." An encounter with successively Benjamin (Walter), Benot de Sainte-Maure, Benton (John), Bernard of Chartres, Bernard of Clairvaux, etc, does not help putting the works in perspective, and to some extent prevents perceiving the degree of completeness of the bibliography. The index is adequate, despite minor mistakes such as "Bertholot" instead of "Berthelot," or "Poiron" instead of "Poirion."
1. E.g.: on p. 78, SJM introduces "hope" in her translation, a singularly clumsy, not to say erroneous interpretation of the adverbial "espoir" of the text, meaning only "maybe."
2. Benedeit, Le Voyage de Saint Brendan. ed., trans., and notes by Ian Short & Brian Merrilees. Paris: Champion Classiques, 2006. The text edited in this volume is the same as the one used by SJM (Manchester University Press, 1979).
3. See, too, an aside on p. 227 when retelling the self-reproaching of Guenevere when she hears the rumor of Lancelot's death: "(Clearly, then, she did not refuse Lancelot because she loves Arthur!)"