Dee’s diaries are layered records: Dee narrates, a scryer reports what he sees and hears, the page assigns speech to figures within the vision, and later notes can revise an earlier identification. Reading those layers separately reveals the action instead of turning every sentence into one undifferentiated supernatural voice.
- Worked action
- 10 March 1582, Mortlake
- Early diary
- Sloane MS 3188
- Later copy
- Ashmole’s Sloane MS 3677
The diary records an event through several filters
The British Library describes Sloane MS 3188 as Dee’s records of messages attributed to angels and other spirits between December 1581 and May 1583. That description identifies the manuscript claim without deciding whether the beings existed independently. Within an entry, Dee supplies the date, place, setup, questions, and retrospective comments. The scryer supplies reported sight and hearing. A name such as VRIEL appears as the label for a speech inside that report. These roles can occupy adjacent lines, so grammatical shifts carry historical information.
The difference is especially clear at the start of Dee’s work with Edward Talbot, the man later known as Edward Kelley. Dee does not write a neutral transcript made by an unseen recorder. He writes himself into the scene, describes Talbot’s posture at the desk, summarizes his own prayer in the oratory, and marks which words Talbot heard. The page is evidence that Dee recorded a mediated exchange. It is not direct access to an unfiltered voice beyond the room.
Worked reading: separate Dee, Kelley, and attributed speech in one dated action
Whitby’s transcript dates the action to Saturday, 10 March 1582, at Mortlake, eleven in the morning. Dee opens in the first person: “One Mr Edward Talbot” came to his house, wanted to see or show something in spiritual practice, and settled before the stone. Dee says that he withdrew to his oratory to pray. This is narrator-level evidence. It establishes who was present, what Dee says each man did, and how Dee framed the purpose of the meeting.
The next layer belongs to Talbot’s reported perception. Within roughly a quarter hour, Dee writes, Talbot had sight of one figure in the stone. Dee then comes to the stone and asks the figure’s name. The record says that the answer was spoken plainly “to the hearing of E.T.” That phrase matters: it assigns hearing to Talbot, even though Dee supplies the written sentence. When the dialogue label changes to VRIEL and the reply says “I am,” the typography identifies the character to whom the diary attributes speech. It does not convert Talbot’s hearing into Dee’s direct audition.
A line-by-line reading therefore produces four claims of different strength. The date and setting are Dee’s written record; Talbot’s presence is Dee’s narration; the sight and hearing are experiences reported through Talbot; and “VRIEL” is the identity assigned within the recorded exchange. The attributed speaker answers questions about Michael and the Book of Soyga, but those answers remain dialogue within Dee’s document. The distinction lets the content be analyzed without either dismissing it as meaningless or repeating its spiritual identity as an independently verified fact.
A later marginal note reopens the scene
At the end of the morning entry, a marginal note states that this was not the true Uriel and points forward to 5 May 1583. The note changes the status of every earlier VRIEL label. It shows that the manuscript itself preserves revision: an identification accepted inside the action was later rejected by Dee or by the annotating hand represented in the transcript. The dialogue is therefore not a stable stenographic record whose labels were certain from the first moment.
The correction also exposes two times on the same page. Event time is 10 March 1582, when Dee recorded Talbot seeing a figure and the voice taking a name. Annotation time is later, after another action supplied grounds for reinterpretation. Reading only the dialogue gives an apparent Uriel interview. Reading the margin with it gives a record of discernment, doubt, and retrospective classification. The contradiction is content: Dee’s project includes procedures for deciding which reported spirits could be trusted, and those decisions could change.
Copies and print can flatten the voices again
Sloane MS 3188 has missing leaves, while Sloane MS 3677 is Ashmole’s later fair copy of the early conferences. A fair copy may preserve material or clarify handwriting, but it is another scribal object. Casaubon’s 1659 True & Faithful Relation adds selection, typography, headings, and a printed reading order. Each stage can make the dialogue look cleaner than the working record and can move a marginal qualification into a footnote, a sentence, or silence.
The dated action supplies a durable method for the rest of the diaries: distinguish the person writing, the person scrying, the figure perceived, the speech attributed to that figure, and any later annotation. Missing leaves limit what survives; later copies may preserve or regularize wording; print supplies another editorial frame. Those layers do not reduce the record to provenance. They reveal how Dee built claims from observation, testimony, dialogue, and subsequent correction within a single continuing investigation.
References
- British Library, Sloane MS 3188 (1581–83 conferences) ↗
- British Library, Sloane MS 3677 (Ashmole copy of early conferences) ↗
- Christopher L. Whitby, John Dee’s Actions with Spirits, volume 2 transcript ↗
- Christopher L. Whitby, John Dee’s Actions with Spirits (University of Birmingham thesis, 1982) ↗
- Folger Shakespeare Library, A True & Faithful Relation (1659) ↗
- Cornell University Library, digitized 1659 True & Faithful Relation ↗