The editorial fortunes of Francesco Petrarca's De vulgarium fragmenta, popularly known as the Canzoniere, is one of the great paradoxes in literary history, as befits the poet best known for his use of oxymora. He lavished attention on the compilation of his vernacular poetry into a lyric sequence and on the transcription of that sequence into book form, as attested to primarily by two manuscripts: the autograph collection of his early drafts now catalogued as Vaticanus latinus 3196; and the partial autograph of the final version, Vaticanus latinus 3195 (while his amanuensis Giovanni Malpaghini transcribed the first part of the sequence, Petrarch had to complete the transcription when the younger man abandoned the job). Thanks to the indications in these meticulous codices and the extensive decoding of them by modern vernacular philologists, it is no exaggeration to say that we know more about the evolution of the Canzoniere than any other early modern work. Indeed it has become a rite of passage in Petrarch studies to recite the various incarnations of the work, particularly in light of Ernst Hatch Wilkins' influential 1955 essay, “The Evolution of the Canzoniere of Petrarch.” Wilkins famously mapped out a trajectory for the sequence through nine progressively larger versions completed over a thirty-year period, from an initial nucleus of at least fourteen poems to a total of 366.

Despite these cares, the philological principles Petrarch had done so much to promulgate, and the immediate popularity of the book, the poet's careful indications were frequently disregarded. In the century after his death it became standard editorial practice in manuscript and the incunabula to adopt separate numbering for the sonnets and canzoni (the common title was in fact I Sonetti e Canzoni di Messer F. Petrarca), to reorder the sequence of the poems, and most fatally, to divide the work into two parts, poems “in vita” and “in morte” of Madonna Laura (the 1525 Vellutello edition went so far as to add a third division, for poems that he could not fit into his ideal chronology). Even when an edition was clearly based on the autograph, as was the case with the 1472 Valdezoco edition printed in Padua, the editors still felt free to amend formal aspects of the text.

The tide began to turn with Mestica's 1896 edition of “Le Rime de Francesco Petrarca restituite nell'ordine e nella lezione del testo originario sugli autografi,” inaugurating a new era in Petrarch bibliography that would come to include prominent critical editions, detailed codicological studies, and, quite recently, a luxury facsimile edition of Vat. Lat. 3195. This prestigious company is now joined by Giuseppe Savoca's new critical edition of the Canzoniere. Published in a two-volume set that includes a monograph reconstructing the internal and external history of the text and describing the editor's critical methods and discoveries (Il Canzoniere di Petrarca tra codicologia ed ecdotica), this new edition sweeps away the accretions of the centuries, in the spirit of today's great art restoration projects.

One of Savoca's most visible restorations is the use of capital letters at the beginning of each verse, in contrast to the prevailing norm of using the lowercase. At the same time he has lowercased various personifications — such as Amore, Fattore, Natura, Morte — that are not capitalized in the autograph. Punctuation is the subject of special scrutiny, and through his truly microscopic examination of the text Savoca discovers thousands of instances of both punctuation marks that were added and of punctuation that was missed. He reduces punctuation to the three signs that Petrarch used — the comma, the period, and the question mark — and eliminates three that were largely unknown before the printing age — the colon, the semicolon, and the exclamation mark. In so doing Savoca breaks radically with all previous editions and produces a text whose music, in his words, is “at times more energetic and syncopated and at others more unstrained and fluid.”

One of the most interesting features of this new edition is the inclusion in the footnotes to each poem of small photographic plates of individual words or phrases from the manuscripts consulted. This innovation gives the reader an unprecedented ability to verify Savoca's readings and to appreciate the exhaustive and meticulous efforts that have gone into the making of this beautiful book.