Anglo-Saxon Futhorc

While Scandinavia reduced its runic alphabet, England and Frisia went the other way. As Old English sound changes split the inherited vowels apart, new runes were added to keep the spelling system working. The result is the futhorc, named for its first six runes, whose sound values had shifted from f-u-þ-a-r-k to f-u-þ-o-r-c.

The Old English Rune Poem lists 29 runes. Northumbrian manuscripts add several more, taking the total to 33 or beyond, though the extra characters are largely scribal rather than epigraphic: they appear in manuscripts about runes more often than in inscriptions cut in wood or stone.

The 29 runes on this site follow the Old English Rune Poem, which runs …lagu, ing, eþel, dæg, ac, æsc, yr, ior, ear. The fullest row surviving in an inscription rather than a manuscript, the Thames scramasax, does not match it: it carries 28 runes, has no ior, and puts ing and dæg ahead of lagu and mann. The missing rune is the least of that. Ior is one of the three characters Page calls pseudo-runes (it appears in the manuscript rows and in no inscription at all), so a 28-rune row is the complete epigraphic futhorc rather than a defective copy of the poem's. That last difference is weaker evidence than it first appears. The seax runs in the familiar order for nineteen runes; its runes 20–24 (ing, dæg, lagu, mann and eþel) then fall into a sequence agreeing with no other surviving row, after which it recovers and runs ac, æsc, yr, ear as expected. Counted against a manuscript row rather than against the poem the disordered stretch comes out as four runes rather than five, because the manuscript rows put dæg ahead of eþel; the positions here are counted against the order this site uses. The same object also has the smith squeezing in a rune he had missed, at the sixteenth position. Page takes the whole irregularity to mean the designer was unfamiliar with runic writing, so this is a garbled row rather than a rival tradition, unlike the Elder Futhark, where several independent witnesses disagree consistently. A position cited against a particular object still has to be counted on that object, which is what this site does.

Where do the English rune names come from? Not, in the first instance, from the Old English Rune Poem, whose stanzas are headed by the runes themselves and never spell the names out. They come from manuscript futhorc rows, chiefly Codex Vindobonensis 795, compiled at Salzburg around AD 798, with ing and æsc taken from Byrhtferth’s Manuscript, where Vindobonensis has the apparently corrupt lug and æs. Cotton Domitian A.ix lists the names too, but almost all of its name-glosses were added in the 16th century by the antiquary Robert Talbot (all but ing and ear), so it is a much later witness than the leaf it sits on. Those two exceptions are not the ing and æsc above, and the pairs are easy to run together: ing and æsc are what Byrhtferth supplies where Vindobonensis is corrupt, while ing and ear are what Talbot found already on the page in Domitian. The names are well attested; they are attested in manuscripts about runes rather than in runes cut on objects, and those are not quite the same tradition.

The English corpus is small but unusually varied: the Franks Casket, the Ruthwell Cross, the Thames scramasax, coins, rings, bone fragments and a good deal of manuscript material. Christian and runic writing coexisted here for centuries: the Ruthwell Cross carries Christian verse in runes, which is a useful corrective to the idea that runes were a pagan-only script.

The Frisian side is smaller again and much less often mentioned. How much smaller depends on who is counting: the old Arntz–Zeiss corpus admitted nine inscriptions, Looijenga counted twenty-one predating the 9th century, and Giliberto’s 2000 edition reaches thirty-one. They are combs, coins, sword fittings, antler and bone pieces, and yew sticks: small portable things, mostly 6th to 8th century, from the coastal Netherlands. The best known are the Arum sword, a miniature yew-wood sword reading edæboda, and the Westeremden yew-stick, which carries the longest Frisian runic text there is.

That handful of objects matters out of proportion to its size, because it is the reason this alphabet is not simply "Anglo-Saxon": the same characters were being used on both sides of the North Sea, and nobody has established which side used them first. The alphabet’s defining innovation (the splitting of the inherited a-rune into ac, æsc and os) had to happen somewhere, and the older assumption that it must have been Frisia, on the grounds that the Continent came first, does not follow from the evidence. The runic finds are early in both places, and the case for an English origin is at least as good. Both readings are still argued, which is why runologists tend to write "Anglo-Frisian" for the innovations and keep the question open.

Two English objects mark out where that argument starts. The Caistor-by-Norwich astragalus, a roe-deer ankle bone from a 5th-century Norfolk cremation urn reading raïhan, is the oldest runic find in England, but its runes are the inherited twenty-four, so it dates the arrival of runic writing rather than the arrival of the futhorc, and it may well be an import. The Undley bracteate, a gold pendant from Suffolk of the later 5th century, is the earliest inscription anyone can argue is futhorc rather than Elder Futhark, on the strength of its æsc rune and probably its os. Neither settles which side of the North Sea innovated; between them they set the earliest date at which anything can be called English runic writing.

The futhorc is the reason English speakers have the letters þ (thorn) and ƿ (wynn) in their earliest manuscripts: both were borrowed out of the runic alphabet into the Latin one.