# Reading the Primary Sources of Art History
> Vasari, Cennini and van Mander are evidence about their own moment before they are evidence about the past. A working method for using them, and the inventories that correct them.
> The Study - 2026-08-02
> Canonical HTML: https://iconotheca.com/study/reading-the-primary-sources-of-art-history/
## Key points
- Vasari's 1568 Giuntina is a different book from the 1550 Torrentiniana, and citing 'Vasari' without naming the edition is an incomplete citation.
- Gaetano Milanesi showed in 1862 that Andrea del Castagno was buried in August 1457 while Domenico Veneziano lived until May 1461, destroying Vasari's murder story after three centuries of repetition.
- The Cennini text in circulation descends from a Laurenziana manuscript of 1437 and later copies, and the first printed edition of 1821 was set from an eighteenth-century transcript rather than the manuscript itself.
- Notarial and probate inventories correct the artist-biography tradition because they were compiled for valuation and division of estates, not for reputation.
## Vasari is a primary source for 1550 and 1568 before he is a source for 1350
## Cennini describes a workshop, and the text you quote is a reconstruction
## Van Mander built a northern canon, and he built it to a programme
## Inventories are the corrective, because nobody compiled them to praise anyone
## A checklist for reading a sixteenth-century account
The primary sources of art history are testimony to be cross-examined rather than foundation to stand on, and the cross-examination is the discipline itself. Vasari's murdered painter outlived his murderer by nearly four years, and it took three centuries for anyone to open the burial register. The register was there the whole time. What changed in 1862 was that Milanesi went and read it, and the habits that produced that correction, which are to go to the document, name the edition and mark clearly what rests on one man's word, are still the only things separating art history from a very old anthology of good stories.
## Questions
**Which edition of Vasari should I cite?**
Cite the one you actually used and name it: the 1550 Torrentiniana or the 1568 Giuntina. Where an argument depends on a passage, check both, because Vasari added, cut and rewrote extensively between them. English translations differ in which edition they follow, so read the translator's preface before treating a passage as evidence for either year.
**Is Cennini reliable for reconstructing fourteenth-century technique?**
He is reliable as a record of what one trained painter held the correct procedures to be, and technical examination of panels corroborates him repeatedly. He is not a survey of practice across Italy, and the text's transmission through a manuscript of 1437 and a sixteenth-century copy means an individual passage should be checked against a critical edition before it carries weight.
**Where can I read a period inventory without going to an archive?**
The Getty Provenance Index publishes transcribed archival inventories of European private collections from 1520 to 1880, and the Montias Database at the Frick Art Research Library covers 1,280 Amsterdam inventories from 1597 to 1681, listing 51,071 works. Both are free to search. Neither is complete, and both say so.
**How should I phrase a claim that rests only on Vasari?**
Attribute it in the sentence, not only in the note, and then say whether any document supports it. If a nineteenth-century editor flagged the passage, name the editor and the year. A reader can then weigh the claim without repeating your research, which is the whole purpose of a footnote.
## Sources
- [1] Discover Arezzo, 'Lives of the most excellent painters, sculptors and architects' (on the 1550 Torrentiniana and 1568 Giuntina editions). Tertiary summary; replace with the printed editions or the Memofonte transcriptions where possible.
- [2] Virtual Uffizi, 'The invention of Domenico Veneziano's murder at the hands of Andrea del Castagno' (Milanesi's 1862 disproof, burial August 1457, Domenico alive May 1461).
- [3] Italian Art Society, 'Florentine painter Andrea di Bartolo di Simone, known as Andrea del Castagno, died suddenly 19 August 1457'.
- [4] Fondazione Memofonte, 'Vasari scrittore' (searchable database of Vasari's writings, both editions of the Vite, Ricordanze and correspondence).
- [5] Treccani, 'Il Libro dell'arte di Cennino Cennini: la prima stampa e il dibattito sulla lingua' (Laurenziana 1437, Riccardiano 2190, Ottoboniano Latino 2874, Palatino 818, Tambroni 1821 from a 1737 copy, Benci's review, Milanesi 1859).
- [6] Studi di Memofonte 15/2015, Alessandro Ricotta on the manuscript tradition of the Libro dell'Arte.
- [7] Archetype Publications, Lara Broecke, 'Cennino Cennini's Il Libro dell'Arte: A new English translation and commentary with Italian transcription' (2015).
- [8] Dictionary of Art Historians, entry for Karel van Mander (Schilder-boeck 1603-04; Miedema's edition and commentary, 1994-99).
- [9] Internet Archive, Karel van Mander, 'Het schilder-boeck', digitised from the Getty Research Institute copy.
- [10] Getty Research Institute, 'What's Covered in the Provenance Index' (2.3 million records; archival inventories 1520-1880; sales catalogues 1650-1945; payments to artists in Rome 1576-1711).
- [11] Frick Art Research Library, 'The Montias Database of 17th Century Dutch Art Inventories' (1,280 inventories, 51,071 works, 1597-1681, Amsterdam city archive; explicit statement of incompleteness).
- [12] Gary Schwartz, '398 Rembrandt's Rembrandts' (the Desolate Boedelskamer inventory completed 26 July 1656 after two days, 363 items or groups of items).
- [13] Rembrandt's Money, 'The Chamber of Abandoned and Insolvent Estates' (cessio bonorum and the role of the Desolate Boedelskamer).