April 17, 2016 update: I have posted two more blogs on runes and Old Norse myths. https://marnietunay2.wordpress.com/2016/04/15/volundarkvida-when-where-and-why/ is the most recent. Previous to that I posted what is the first part of a two-part introduction to runes and John G. Bennett’s theory of triads of will: https://marnietunay2.wordpress.com/2015/11/02/introduction-to-runes-and-triads/ In that first part, I tried to give people with little or no background in either runes or triads theory the sense that there might be a valid corresponding system of correspondences between the two. I didn’t worry too much about the traditional ordering of the runes, as I wanted to present the most persuasive data suggestive of a link first. Part two, which is next up on my bucket list of blogs, will address the rest of the runes in the traditional elder futhark, as well as go more deeply into the two runes I touched on only briefly in part one of the introduction, as the case for them is more complex. After that, I intend to address issues connected with the hands-down most-discussed of the Old Norse Eddic poems, namely, Völuspá, beginning with a good look at the stanzas relating to Gullveig and the witch Heiði: http://www.voluspa.org/voluspa21-25.htm
ONE: THE ORIGINS OF THE ELDER RUNIC SCRIPT
“All authors who have contributed to the genetic problem have been led sooner or later to postulate the intervention of an individual creator (or, alternatively, of a small group). (1) It appears, indeed, that the runic alphabet is based on a careful phonological analysis of the OGmc. sounds, rather than on haphazard borrowing. (2) Such a creative act would of course be far less bound to external circumstances than the diffusion of an ornamental fashion.” [Derolez (1) p. xvi]
In his note to the above remarks, Derolez quotes with approval Bruce Dickin’s ““appraisal of the Old English runic character” : … “a cursory glance at Caedmon’s Hymn or Bede’s Death-song will show how vastly superior as an instrument for recording the sounds of Old English it is to the latin [sic] alphabet, which had eventually to borrow from it for use in England the characters ‘þ’ and ‘w.’”” [ibid, n. 2]
“…all serious modern students of the runes have recognized the fact that runic writing is derived from the great Mediterranean writing tradition. Their greatest efforts have been directed toward the identification of a particular Mediterranean alphabet as the source from which the runes were borrowed…“ [Morris, p. 3]
I. Richard Morris’s theory of Archaic Greek Origins: After a detailed survey of the major theories regarding the origins of the runes that earned high praise even from his scholarly adversary Bengt Odenstedt (1), Morris set out his reas-ons for favoring the archaic Greek ca. “5th – 4th century before Christ” [Morris, p. 154] over the Latin alphabets “ca. 4th – 3rd century before Christ” [ibid] as the most likely source for the runes, some of the more compelling of which I will summarize here:
“Latin had no letters for /p, z.w.j/ which were necessary for the Germanic sound system.” [Morris, p. 152] Greek had letters for the z and w; and, the argument against using Gk. heta on phonological grounds for the H-rune did not apply 500 B.C. [ibid, p. 3]
Greek koppa Ϙ could have been used to represent the sequence /ng/ in Germanic, i.e. the rune. [ibid, p. 153]
In his table showing “the runes and the archaic Greek and Latin alphabets,” [ibid, p. 154] Morris shows that, with respect to one of the thornier issues in runic origins theory, there is a ‘th’ sound represented by a single letter in the Greek, namely, the archaic theta, that corresponds to that in the Elder Futhark rune þ ; however, there is no such correspond-ing letter in the Latin alphabet.
The archaic Greek also had a ‘z’ sound corresponding to what is thought to have been the original sound of the Elder Futhark rune , whereas the Latin of that period did not. [ibid]`
The forms Ι and of Greek iota could have been used to form the runic graphs , , and even modified to create the runic rune , æ . [ibid, p. 153] There is no such corresponding possibility for the development of the rune from the Latin alphabet. [ibid, p. 154]
Elmer Antonsen, who also favored the archaic Greek theory of runic origins, argued that “the presence of the 6th vowel-rune , , in the fuþark” can be explained “on the basis of linguistic evidence;” saying, “….it is apparent that the runes and originally represented only short vowels, since their names begin with short vowels. Had they also been the designations for long vowels, their names would have begun with a long vowel, as is true of the runes and , and is still true for letters of the alphabet in the West European tradition… There is therefore an exact fit between the vowel system of Proto-Germanic (2) and the orthographic system provided by the fuþark. This must mean that runic writing arose during the Proto-Germanic period (2), that is, well before our earliest runic records.” [Antonsen, p. 46 – 47]
Martin Findell remarks that “The absence of syllabic discrimination in the use of Roman ˂i˃ ~ ˂j˃ is sufficiently well known to require no further comment. On the other hand, the fact that the fuþark contains distinct runes i/j and u/w suggests that to speakers of early Gmc dialects (or at least, to the creators of the fuþark) the distinction was perceived as being significant.” [Findell, p. 54]
Later, Findell concludes: “It appears that in the Continental runic inscription, as in Older Fuþark inscriptions elsewhere, the “yew rune”  is functionally a free variant of i… The choice of this rune does not appear to be motivated by the phonetic environment… In the two reliable cases where [the original value of ] appears, we have good grounds for sup-posing it represents a high front vowel. We have no basis for arguing that the selection of Ï is motivated by any phonetic or phonological distinction – it does not represent a distinct allophone of /i/ or Ï. What, then, is the motivation for its selection? Why use a rare and phonologically redundant rune when a more common and formally simpler one is avail-able? This must remain – for the time being, at least – an open question.” [Findell, p. 218]
Michael Barnes does not dismiss Antonsen’s idea of an archaic Greek genesis outright. He says: “The case that leads Antonsen (and his pupil Morris — cf. Morris 1988) to move the origin of runic writing back to near the middle of the first millennium B.C. is cogently ar-gued and by no means without interest.” However, “It founders not because of its artifi-ciality but because of the dearth of runic inscriptions between the supposed period of origin and A.D. c. 160, and the lack of a trail of such inscriptions leading from the Mediter-ranean northwards towards Scandinav-ia.” And, “My complaint against Antonsen and Morris is rather that they have been unwilling to engage with the reasonable criticisms their proposal about the high age of the fuþark has attracted.” [Barnes (2), p. 24]
(Antonsen goes on to set out “yet further evidence pointing to the origin of runic writing in the Proto-Germanic period” related to linguistic changes, which I will discuss later in this appendix.)
In his detailed discussion of Morris’s archaic Greek theory of runic origins, Odenstedt says: “To sum up: I find Morris’s theory unlikely chiefly for chronological reasons. If the futhark had been invented as early as 500 BC, we would no doubt have had runic inscrip-tions on the numerous early Iron Age objects that have come down to us…” [Odenstedt (1), p. 382]
Earlier on in his 1990 discussion of Morris’s book, Odenstedt agrees that the Meldorf fibula “might justify Morris’s opinion that the invention of the futhark must be pushed back considerably.” [ibid, p. 364] If only it were runic. But Odenstedt does not agree with Morris and Antonsen that the Meldorf fibula is in fact runic.
Whatever one thinks of the various arguments made for or against a positive identification of the Meldorf fibula as being runic, (or Latin, for that matter), I agree with Bengt Oden-stedt that it’s a very dodgy hook on which to hang a whole theory.
II. Hendrik Williams’ version of the Latin theory, the ‘Switched at birth’ hypothesis: Bengt Odenstedt is a proponent of the Latin theory of runic origins, and so is Henrik Williams, who developed his own version of the Romans-on-the-Rhine theory, the ‘switched at birth’ variation:
“The Germanic language had all the phonemes of Latin but also six other phonemes (or perceived phonemes), /w/, /ŋ /, /j/, /ï /, /z/ and /d/… According to my derivation all of the five “unnecessary” Roman letters [K, Q, X, Y, Z) were used to make runes for these for these extra sixth Germanic phonemes; one more rune was needed for the sixth extra phon-eme, and therefore invented ( probably by the doubling of D). “ [Williams, p. 216]
To get around the obvious discrepancies between his “proto-runes” and “the actually attested sound-values of the fuþark,” Williams postulates that “all the proto-runes with novel i.e. Germanic sound values have “been switched,” or “traded places,” or “been mixed-up…” When the traditional alphabet order was rearranged into the fuþark se-quence, there was probably a mix-up of certain graphemes, whereby five or six sets of runes were switched pairwise.” [ibid]
III. Flashpoints in Discussions of Runic Origins
IIIA. The Ing-rune: One of the major flash-points in scholarly debate on runic origins is the question of the ‘Ing-rune,’ or .
Terje Spurkland briefly summarizes the issue for scholars with respect to this rune: “This sign seems redundant because rune-carvers could use ng just as well. In fact, there are instances where and are used indiscriminately in inscriptions in the elder fuþark. On the Reistad stone in Vest-Agder in southern Norway, we find iuþingaR , which is interpreted as the name of a dead person. On the stone from Arstad in Rogaland, we might be able to read R, ‘young friend.’ From these examples we can see that the sequence /ng/ is written both as and as .” [Spurkland, p. 7]
To this, Michael Barnes adds that “[The Ing Rune] cannot in point of fact tell us anything about the age of the runes, since it stands either for/ŋ/, which was a variant of /n/ that occurred immediately before [g] (as English single versus sinful), or for [ŋ] or even [iŋ] + [g] a quicker alternative to writing [ ].” [Barnes (1), p. 13]
However, Elmer Antonsen argued that “The most conspicuous of all symbols in the older runic alphabet is the rune , which is often transliterated as ŋ in spite of the fact that there was no phoneme */ ŋ / in the older Germanic languages… The sequences */ng/ and */nk/ were pronounced / ŋg/ and / ŋ k/ (as in English longer and honker.) The velar nasal in these sequences was merely an allophonic variant of */n/ and as such would not be represented in an alphabetic writing system. The ng-rune does not represent this velar nasal, but rather the cluster of nasal + stop. Since this rune, and only this rune, represents a phonemic cluster, it does not fit well into the alphabetic system… No satisfactory ex-planation of the rune has yet been put forward, but one thing is clear: this rune must have its origin in the Greek tradition, for only there do we find a special designation for /ng/, although nasals are not ordinarily designated before consonants. The ng-rune must somehow be connected with Greek gg…” [Antonsen, p. 102 – 103]
In his review of Richard Morris’s book, Bengt Odenstedt doesn’t have much to add to the discussion on the Ing rune , except to: (i) “agree with Morris and Antonsen that. are to be regarded as bind-runes: + …. ” [Odenstedt (1), p. 373]; and, (ii) essentially, to re-state the sole point [Odenstedt (2), p. 163] where his own ‘Latin theory’ of runic origins differs from that of Fritz Askeberg’s ‘Latin theory:’ “It is not necessary to go to preclassical Greek Ϙ quoppa [as Richard Morris does] to find a model for . Classical Latin Q (which derives from quoppa) provides an equally plausible model. The new sound value in the futhark is not so strange, considering that Germanic people had no need for Q with its Latin sound value: they therefore used it to denote a (homorganic) nasal that had no letter of its own in Latin…” [Odenstedt (1), p. 377]
However, in his book, ‘On the Origin and Early History of the Runic Script…,’ Odenstedt gives a protracted discussion on the Ing rune. [Odenstedt (2), p. 103 – 118, 163, 167]
Speaking in general about the sound value of the Ing rune, Odenstedt remarks that: “It has often been stated by runologists that was a superfluous rune since it would have been possible to use n + g (see e.g. Grønvik 1981:28f.), but since at this time g was pro-nounced Ʒ [sic but see also Note (3)] everywhere except in the combination ŋg… it is pos-sible that ŋg was perceived as one phoneme, contrasting with e.g. n; if so is of course a necessary rune…” [Odenstedt (2), p. 167]
Odenstedt accepts Fritz Askeberg’s theory Latin-derived theory of runic origins (4) – except insofar as it pertains to the Ing rune: “As I have stated above, I find Askeberg’s derivations convincing, except on one point: the ŋ-rune. As I have shown (16. a–c), the original form of the ŋ-rune was probably . As we shall see below, this rune should no doubt be derived from Roman Q, the only letter in the Roman alphabet that Askeberg does not use in his derivations…” [Odenstedt (2), p. 163].
The first problem with Odenstedt’s assertion is that the data he provides in the section he refers to (16. a–c) does not support at all his premise that the original form of the Ing rune was . In fact, even in his very own table of the earliest known examples, the does not show up until the 6th c. in the Vadstena and Motala inscriptions. [ibid, p. 107] Meanwhile, even he agrees that the Kylver inscription is a “certain” example of the Ing rune, and Kyl-ver shows the form . Another 5th c. form given in Odensted’s table is that in the Opedal inscription:
taken from: http://www.arild-hauge.com/arild-hauge/no-rune-opedal.jpg for more pictures of runic inscriptions, see also: http://www.arild-hauge.com/norway.htm
and the Opedal inscription is the only example given by Odenstedt of that form in his table dating from the 5th c. through the 6th c. Even if the Opedal inscription is assumed to be a rather smushy variation of the , it’s hardly a solid support for Odenstedt’s assertion regarding the original form, given that his own table also shows a “certain” 5th c. example of the form, in the Szabadbattyán inscription, as well as two more “certain examples of the same form that he dates “c. 500,” namely the Kong and Tanem inscriptions, as well as one “certain” 6th c, inscription, Aquincum and one “uncertain” 6th c. inscription, Grumpan – and all of this he finds wanting against two mere 6th c. examples, Vadstena and Motala, of the form !!. So how does Odenstedt justify his assertion that is the original form of the Ing rune? He does this by casually and quietly equating it with the form given in the Kylver inscription: “… In the five inscriptions in which the sound value of the [Ing] rune can be determined none exhibits the sound value / ŋ /. But in the three futhark inscriptions in which the rune indisputably occurs (Vadstena, Motala, Kylver) it has the form …” [Odenstedt (2), p. 108]. And that’s how the trick is done. Odenstedt equates the 6th c. form a 5th c. form, because the 6th c. form fits a lot more conveniently into his theory regarding the Latin Q origin of the Ing rune than the actual 5th c. forms and would do!
IIIB. The ‘Peorð’ Rune: On the other hand, I find Ray Page’s 19994 assault on Odenstedt’s theory, in Page’s essay ‘Quondam Et Futurus,’ to be harsh and unfair, at least with respect to the ‘P’ rune. Page rips Odenstedt (justifiably) for not including St. Cuthbert’s coffin and the so-called Pada coins in his survey of runic inscriptions [Page (1), p. 10 – 11]. Fair enough, but then Page goes on to say: “I think it is beyond the bounds of probability to assert there is no continuity between the earliest use of this graph [the rune-form], its continued use into the sixth century in fuþarks outside England, and its appearance on the Pada coins and thereafter. But by marginalising the coin evidence and ignoring that of the Cuthbert coffin, Odenstedt is enabled to deny the likelihood that ‘p’ would be used in [the] Watch-field [inscription]. But by marginalising the coin evidence and ignoring that of the Cuth-bert coffin, Odenstedt is enabled to deny the likelihood that ‘p’ would be used in Watch-field. What slight evidence there is has been skewed to fit a hypothesis.” [ibid, p. 11]
But, here is what Odenstedt actually says: “11.4.1 As we have seen, seems to have been avoided in runic texts in Scandinavia and on the continent; the only examples occur in futhark inscriptions. On this point runic practice in England was different – was used to indicate p in runic words, at least in the later inscriptions. In my material there occurs one example, helle ‘help’ (subjunctive) in 168 Mortain (c.750). That was a rune in liv-ing use in England is confirmed by other inscriptions which fall outside my material. The Whitby comb uses it in the word h[e]lipæ ‘help’ (Page 1973:168 and Plate 4), and it is also used on the pada and epa coins (ibid:123ff. and Figures 18,19) The use of in these late inscriptions seems to be an English innovation. There is some reason to believe this in-novation is comparatively late. To judge from a recently discovered sixth-century inscrip-tion, 152 Watchfield (c. 525 – 550), was still avoided in words containing p, just as in Scandinavia and on the continent.” [Odenstedt (2), p. 80 – 81]. Odenstedt goes to discuss the Watchfield inscription “at some length” and concludes that it may well be “a case where the rune is used to indicate the rare consonant p.” [ibid, p. 81]
I wouldn’t call that a case of “‘denying the likelihood’ that ‘p’ would be used in Watchfield.” The fact is, that Odenstedt’s case for making out the rune in Watchfield as actually being a stand-in for the sound designated by the rune in the Elder Futhark makes a lot of sense; so he is correct in saying that there is “some reason” to believe the use of the latter rune in runic inscriptions (as distinct from fuþark-rows] may have come late to England.
NOTES (1) “Morris’s account of previous theories of the origin of runes and of early Greek and Latin epigraphy is highly interesting. Those who want an introduction to these subjects can hardly find a better guide than Morris.” [Odensted (1), p. 360] “… this is a well-written, well argued, knowledgeable and stimulating book.” [ibid, p. 358] “Chapters 2 and 3 (p. 55 – 105) are devoted to a thorough discussion of pre-classical Greek and Latin alphabets. In my opinion these two chapters are the best in the book and invaluable for those who want to study the origin of runes in the future…” [ibid, p. 367]
(2) Antonsen doesn’t think to define what he means by ‘Proto-Germanic,’ which is really unfortunate, because “Estimates of when Proto-Germanic became a distinct language vary considerably. There is general agreement, however, that, by the beginning of the Nordic Iron Age around 600 B.C.E, the inhabitants of southern Scandinavia were speaking a language that can reasonably be termed Proto-Germanic.” [Sheffield, p. 4]
(3) Odenstedt means ‘like insular g‘ but there seems to be some disagreement as to how that should be represented; see for example: http://en.wikipedia.org/wiki/Ezh#Ezh_and_yogh
April 25, 2015 NOTE: I fully intend to continue to develop this appendix on runic origins over the next year, but I’ve become bogged down in the complexity of the data and the arguments, and I need to move on right now.
TWO: OLD NORSE MYTHS and RUNIC SOURCES: Summary Reviews and recommendations
I’ve launched this second appendix as a separate blog-post:
as it will be updated fairly regularly over the next year, and scrolling down on this page will get annoying.
Sources: Antonsen, Elmer H. ‘Runes and Germanic Linguistics.’ Mouton de Gruyter. 2002 Barnes, Michael P. (1) ‘Runes: a Handbook.’ The Boydell Press, 2012 Barnes, Michael P. (2)., ‘What is Runology and Where Does It Stand Today?‘ in ‘Futhark: International Journal of Runic Studies 4 (2013)’ http://www.futhark-journal.com/issues/ Derolez, René. (1). ‘Runica Manuscripta.’ De Tempel, Tempelhof 37, Brugge. 1954 Findell, Martin. ‘Phonological Evidence from the Continental Runic Inscriptions.’ De Gruyter. 2012 Morris, Richard L. “Runic and Mediterranean Epigrahpy.” ‘NOWELE, Supplement Vol. 4,’ Odense University Press. 1988
Odenstedt, Bengt, (1) ‘A New Theory of the Origin of the Runic Script: Richard L. Morris’s Book Runic and Mediterranean Epigraphy,’ in Bammesberger, Alfred, ed., ‘Old English Runes and their Continental Background.’ Heidelberg. 1991. p. 359 – 387 Odenstedt, Bengt, (2) ‘On the Origin and Early History of the Early Runic Script: Typology and Graphic Variation in the Older Futhark.‘ Acta Academiae Regiae Gustavi Adlophi LIX, Uppsala. 1990 Page, R.I. (1). ‘Runes and Runic Inscriptions, Collected Essays on Anglo-Saxon and Viking Runes.’ The Boydell Press, 1998 Robertson, John. S. ‘How the Germanic Futhark came from the Roman Alphabet.’ http://www.diva-portal.org/smash/get/diva2:512974/FULLTEXT01.pdf published in Futhark: International Journal of Runic Studies Vol. 2 2011 http://uu.diva-portal.org/smash/get/diva2:512970/FULLTEXT02.pdf
Sheffield, Ann Groa, ‘Long Branches, Runes of the Younger Futhark,’ Lulu, 2013 Spurkland, Terje. ‘Norwegian Runes and Runic Inscriptions.’ Translated by Betsy van der Hoek. The Boydell Press. 2005 Williams, Henrik (1). ‘The Origin of the Runes.’ in ‘Frisian Runes and Neighboring Traditions.’ 1996, p. 211 – 218.