How does "handwriting" provide evidence for dating Beowulf's composition to "the first half of the eighth century"?

Question

The Wikipedia article for Beowulf has this interesting bit in the "Authorship and date" section:

On the other hand, some scholars argue that linguistic, palaeographical (handwriting), metrical (poetic structure), and onomastic (naming) considerations align to support a date of composition in the first half of the eighth century; in particular, the poem's apparent observation of etymological vowel-length distinctions in unstressed syllables (described by Kaluza's law) has been thought to demonstrate a date of composition prior to the earlier ninth century.

How can handwriting help date the composition of the poem? The same article notes that there is only a single surviving manuscript, created 975–1025, and transcribed by two copyists who had noticeably differing handwriting. Thus I'm confused how handwriting by later scribes could provide evidence for an earlier date.

The three citations offered for the line in question consist two books and a paywalled article; I cannot access them. Attempted searches led to discussion of dating the manuscript with the handwriting (which makes more sense than dating the composition) and a lot of arguing for other dates, or arguing for similar dates but with non-handwriting reasons.

How does "handwriting" provide evidence for dating Beowulf's composition to "the first half of the eighth century"? What reasoning/analysis did the "some scholars" in my Wikipedia quote use?

---

This question is not asking if the handwriting in question is good evidence, or even evidence at all, for the scholars' claims. It is not asking when Beowulf was composed. I am uninterested in rehashing those arguments. All this question asks is what the evidence for the specific claim noted is.

Answer 1

The history of a Wikipedia article can be investigated using the ‘wikiblame’ tool. This shows that the gloss “(handwriting)” for “palaeographical” was added, along with a couple of other glosses, in this January 2021 edit by ‘Chiswick Chap’. This kind of gloss can be helpful in smoothing the path for readers (avoiding the need to click on the link to the “palaeography” article), but it risks giving a misleading simplification. In this case the problem is that the term “handwriting” suggests that the evidence referred to is the style of individual scribes, but palaeography is about more than that.

Before the 2021 edit the sentence read:

On the other hand, some scholars argue that linguistic, palaeographical, metrical, and onomastic considerations align to support a date of composition in the first half of the eighth century

This sentence was added, with essentially the same wording, in this 2014 edit by ‘Altenmaeren’, and cited to Leonard Neidorf’s The Dating of Beowulf: A Reassessment, unfortunately with no page number. However, it does seem to be a fair (if very compressed) summary of the cited work:

The contributors reassess the chronological implications of a wide variety of evidence, including, but not limited to, the linguistic, metrical, semantic, onomastic, paleographical, cultural, historical, and theological.

_{Leonard Neidorf, ed. (2014). The Dating of Beowulf: A Reassessment, p. 17. Cambridge: D. S. Brewer.}

What this summary omits is any explanation as to how palaeographical evidence can contribute to an eighth century date for the composition of Beowulf, when everyone concerned agrees with an 11th century date for the extant manuscript. For the explanation, we have to read as far as chapter 12 in Neidorf, where we find:

Lapidge argues that a set of “literal confusions” (miswriting one letter for another) in the surviving MS of Beowulf arise from the difficulties later scribes had in interpreting letterforms in an archetype written before 750.

_{George Clark (2014). ‘Scandals in Toronto: Kaluza’s Law and Transliteration Errors’. In Neidorf, ed. (2014), pp. 227–228.}

Michael Lapidge had examined errors in the Beowulf manuscript that probably occurred because the scribe had trouble reading the corresponding letter in the exemplar they were copying. For example, he gives three cases where the scribe probably confused the letters “a” and “u”:

1. “unhar” for “anhar” (singularly hoary) in line 357 (eald ond anhar mid his eorla gedriht)
2. “wudu” for “wadu” (depths) in line 581 (wadu weallendu. No ic wiht fram þe)
3. “strade” for “strude” (plunder) in line 3073 (Næs ða on hlytme, hwa þæt hord strude)

and then comments:

Let us now consider the palaeographical implications of such confusion. The letter a is one letter which changed its form in the various systems and grades of minuscule script; it may therefore serve as a useful diagnostic instrument in our attempt to identify the script-type which gave rise to the errors of transliteration I have recorded above.

_{Michael Lapidge (2000). ‘The Archetype of Beowulf’, p. 12. Anglo-Saxon England 29.}

Lapidge says that this confusion is not possible in 10th century “square” minuscule, nor in 8th and 9th century “hybrid” minuscule, but going back to the “cursive” variant of “set” minuscule of the early 8th century:

Among the letter-forms which are characteristic of Anglo-Saxon cursive minuscule is what palaeographers call an ‘open’ a: that is, an a written in a single, horizontal-tending stroke, as follows:

It is obvious that the open a of cursive script is the letter-form which could most easily be confused with the letter u

_{Lapidge, pp. 16–17.}

By itself, this is not conclusive (anyone writing quickly might use an open “a” occasionally), but Lapidge considers four other confusable letter-pairs: “r” and “n”, “p” and “ƿ” (wynn), “c” and “t”, and “d” and “ð” (eth), and concludes:

the very substantial number of literal errors in the transmitted text of Beowulf can most economically be explained, and at a stroke, by the supposition of an early-eighth-century exemplar in set minuscule.

_{Lapidge, p. 35.}

So, returning to Wikipedia, it seems that the editor ‘Chiswick Chap’ who added the “(handwriting)” gloss had not read Lapridge’s palaeographical argument, otherwise they would have realised that the gloss was inadequate to convey the way in which handwriting evidence contributes to the early date of composition. I suggest you be bold and edit the article!