The Rightward Slant in the Handwriting of the 17th/18th Century Scribe of To Theodore

For years I took Agamemnon Tselikas at his word on one very specific, very simple question:

“Are there any handwriting samples from Mar Saba or the Patriarchate Library that resemble the Letter to Theodore?”

I didn’t ask this once. I asked it over and over—easily ten times—through a mutual friend. And ten times the answer came back: no. Nothing really comparable. No close matches.

At the time, I didn’t know anything about the details of Quesnell’s requests to Dourvas or about the specific comparanda that had actually been pulled from Mar Saba and shown around. I just had the repeated negative: “no, there’s nothing like it.”

And then, years later, Tselikas publishes the claim that the Themata manuscripts look very much like the letter of To Theodore.

That was the moment the alarm bells started ringing.

My other friend, Charlie Hedrick, noticed the same thing in person. When he met with Tselikas, he realized that Tselikas had just recently published something on the Themata manuscripts. The whole move suddenly looked a lot less like a deep palaeographic insight and a lot more like the professor grabbing whatever he had most recently worked on and pressing it into service as “the model” for To Theodore. In other words: a lazy choice, not a real similarity.

Now that AI has caught up enough to be visually useful, I decided to put this to the test in a way that simply wasn’t available to Quesnell or to Smith. I asked ChatGPT 5.1 a straightforward question:

Which is more similar to the manuscript of To Theodore from Mar Saba: the manuscripts Dourvas showed Quesnell, or the Themata manuscripts that Tselikas proposed?

The short answer
On balance, the Mar Saba pages of To Theodore look a lot more like the “Dourvas comparanda” than like the Themata manuscripts Tselikas used.

That doesn’t prove forgery or authenticity; it doesn’t identify a scribe. But it absolutely undercuts the idea that the Themata manuscripts are the “natural” or “obvious” model for the Mar Saba hand. They simply are not.

A bit more detail

If we stay at the level of broad palaeographic features—script type, slant, ductus, letter-forms—without wandering into authorship claims, the pattern is pretty clear.

1. General style and ductus

To Theodore + Dourvas pages
The Mar Saba letter and the Dourvas exemplars share the same sort of everyday 17th/18th-century monastic cursive:

• Medium-fine pen, not the heavy, formal book-hand one sees in high-end liturgical codices.

• Fairly loose letter-shapes: the hand looks written, not drawn, with a rhythm that carries through the line.

• A moderate rightward lean of individual letters, but not so exaggerated that it looks theatrical.

• Lines that wander just enough to tell you this is real workaday writing on the page, not a calligraphic showpiece.

In short, they look like the kind of script a working monk or scribe would use day in, day out.

The Themata manuscripts
By contrast, the Themata manuscripts that have been put forward as “very similar” to To Theodore belong to a noticeably more formal and “bookish” script profile:

• Heavier pen-stroke and more ink on the page.

• More dense ligaturing: letters hooked and fused into each other in a characteristic late-Byzantine way.

• More ornamental abbreviations and standardization of forms.

They look like something you might imitate, yes—but they do not look like the same style as To Theodore. They are a different register of script.

2. Slant and line behaviour

To Theodore + Dourvas pages
In the Mar Saba letter, the individual letters lean mildly to moderately to the right, while the baselines themselves remain relatively stable. This combination—right-leaning letters on lines that don’t dramatically fall or rise—is also visible in the Dourvas pages.

Themata pages
The Themata samples typically show a different balance:

• More line-to-line variation in angle.

• A different interplay between vertical and diagonal strokes, especially in kappa, mu, nu, and omega.

• A feel that is, again, more formal and “tight” than the Mar Saba hand.

3. Letter-forms and ligatures

In To Theodore, the core minuscule letter-forms—alpha, epsilon, eta, omega—are relatively simple and open. Ligatures are present, of course (this is post-Byzantine Greek), but they’re not deployed with the heavy, ornamental density you see in more bookish hands.

The Dourvas exemplars sit in the same stylistic space:

• The basic shapes of alpha, epsilon, eta, omega line up closely with the Mar Saba forms.

• The way letters are joined, especially in running text, feels very “of a piece” with To Theodore: efficient, legible, not fussed-over.

The Themata hands, on the other hand, are more compressed and “tight”:

• Letters pack more closely together.

• There is a heavier reliance on inherited ligature habits which To Theodore simply doesn’t use as often or in the same way.

Again, these differences don’t magically trace an author’s identity, but they do map out script families. And To Theodore simply belongs with the Dourvas-style, workaday monastic cursive, not with the more formal Themata hands.

What this does not show

It’s important to be explicit about what this kind of comparison does not prove:

• It does not show that the same person wrote To Theodore and the Dourvas samples.

• It does not rule out that the Themata hands (or similar formal models) might have influenced someone at one or two removes.

What it does show is that, at the level of overall script type, To Theodore sits comfortably inside the same 17th/18th-century Patriarchal/monastic cursive family as the Dourvas examples—and it does not particularly resemble the specific Themata hands that have been advertised as its “twins.”

Or, to put it more bluntly: if you ask, “Which set does the Mar Saba letter resemble more, Themata or Dourvas’ Mar Saba examples?”, the evidence points clearly toward the Dourvas examples.

Which specific Dourvas page is closest?

When pressed to choose among five different images from the Dourvas set, the closest match to To Theodore is the third image (page 17).

Why that one?

• The overall ductus—the speed and looseness, the slightly messy-but-consistent stroke quality—is very close to the Mar Saba letter.

• The rightward slant is almost exactly in the same range as To Theodore: more pronounced than in the first two pages, but not as stiff or upright as the later ones.

• The size of the script, word-spacing, and line-density are all in the same ballpark as To Theodore—moderately small hands, fairly tight lines, minimal ornament.

• Several specific letter-forms in the body text—especially alpha, epsilon, eta, omega—are shaped and joined in ways very reminiscent of the Mar Saba hand.

The fourth image (page 18) isn’t bad as a runner-up, but if one has to pick the single closest match, image 3 / page 17 wins.

What this says about expertise and judgment

None of this is about “owning” an expert. Tselikas is, by any reasonable standard, a highly qualified palaeographer of post-Byzantine manuscripts. But expertise does not make someone infallible—especially when they are under pressure to produce quick comparanda in a controversial case.

If he tells me, over and over, that there is nothing in the Mar Saba or Patriarchate collections that looks like To Theodore, and then later discovers “very similar” material in Themata manuscripts he already happened to be working on, I have to ask: was that a fresh insight, or a convenient retrofit?

The AI comparisons don’t solve the Letter to Theodore debate. What they do is quietly but firmly remove one of the props that has been holding up a particular narrative: the idea that Themata provides the natural handwriting model for the Mar Saba letter.

On the level of script type and broad palaeographic features, that claim just doesn’t survive contact with the evidence.