In his most recent commentary on the ambiguities and riches of Saussure's thought, Michel Arrivé aims to describe its genesis and development by reference to all of the writings to which we now have access. It is not always remembered that, the early publication of the Mémoire and the thesis apart, the chronology of publication of Saussure's work is almost inverse to the order in which it was produced. Starting with a ‘bio-chronology’, Arrivé reminds us that from the early years of his academic career until shortly before his death Saussure was profoundly and simultaneously involved in work in the three areas of linguistics, legend and Latin anagrams, which dispels the idea of a Saussure who accomplished relatively little during his years at the University of Geneva. This tripartite division – Arrivé is not much interested in Saussure the diachronic linguist – recurs throughout this volume, as it is by drawing together the threads from all of Saussure's work that Arrivé sheds light on the inherently interesting relationship between these three areas.

The careful re-reading of the Cours de linguistique générale (CLG) in the second chapter allows Arrivé to challenge certain fixed ideas, and to formulate a number of ideas which will be examined in greater depth in subsequent chapters. He analyses the problems that arise from the fact that the term parole is polysemic, even within the CLG. That one of its meanings is equivalent to énonciation fits well with the references to the sujet parlant (often overlooked simply because the term does not figure in the index to the Cours). The discrepancy between a ‘linear’ linguistic sign and clearly non-linear units in other semiological systems has long troubled some critics in the debate about whether or not the archetypal sign is linguistic. Here it is once again Saussure's attempt to work towards a satisfactory terminology for linguistics, combined with certain decisions by the editors of the CLG, that obscures the fact that it is the significant which must necessarily be linear for langue to function as a system, and not the signifié or the sign – although of course certain other problems occur if they are distinguished in this way, as Hjelmslev noted. (Chapter 3 of Arrivé's book will return to semiology, and chapter 8 will look afresh at the debt of early Greimas and early Barthes to Saussure.)

In chapter 7, Arrivé goes on to show that Saussure would have struggled to reconcile the non-linear nature of the significant with his own interpretation of the anagrams, which depends on the play of non-linear items, and suggests that for Saussure the latter is the essence of littérarité. Putting together the various writings, we can see that for Saussure literary conventions have a fixed character, unlike semiological systems whose fundamental characteristic is to evolve. ‘Legend’ is located somewhere between the two; in langue, of course, the signs are limitless, whereas in legend they are not, and we are dealing with what Saussure calls ‘un symbole indépendant’ in his ‘Notes pour un article sur Whitney’ (a text which Arrivé finds to be one of the richest in the Écrits de linguistique générale). All of this explains what some see as the surprising omissions of literature and legend from the list of semiological systems: despite his claims for the centrality of characteristics of langue for the study of other sign systems, Saussure was in fact more hesitant than some of his followers to extend one form of analysis to a multitude of disciplines.

In an interesting chapter, Arrivé deals with one of the areas that he has written on most, i.e. the idea of the Unconscious in Saussure. He touches briefly on the rendezvous manqué of Saussure and Freud, before suggesting that various levels of consciousness are postulated by Saussure. He also sets the record straight by showing that for Saussure speakers are unconscious of linguistic units in themselves, being conscious only of differences, although Lacan claims for Saussure the reverse. Finally, to refer to just one more of the persistent idées fixes, Arrivé demonstrates that the notion that Saussure was not concerned with syntax, in part promoted by Chomsky, overlooks the fact that syntagme for Saussure has a much wider meaning than the word now has for us, as evidenced by a passage in Riedlinger's course notes: ‘cette notion de syntagme peut s’appliquer à des unités de n’importe quelle grandeur, de n’importe quelle espèce’, a comment which becomes even more significant in the context of the view of the dynamic acte de langage that emerges from the Écrits de linguistique générale.

Many of the intriguing dichotomies and frustrating lacunae remain (inevitably so in an author who, as it is sometimes said, did not write what was published and did not publish what he wrote). It is impossible in a review to do justice to the rich diversity of chapters in this volume (which includes a playful pastiche of a note inédite), but thanks to Arrivé's book, Saussure's thought is now better elucidated and capable of being better understood than before.