Purveyors of the Magic of Imagination
Llyfrawr: Home of the Mathomium
Tolkien and Welsh
From the Preface

Tolkien and Welsh provides an overview of J.R.R. Tolkien’s use of Welsh in the nomenclature of his Legendarium, ranging from the obvious (Gwynfa), to the apparent (Took), to the veiled (Gerontius), to the hidden (Goldberry).

Though Tolkien and Welsh is a book by a linguist, about Tolkien’s linguistic creations, it was written for the non-linguist with the goal of making the topic accessible to a larger audience. The unavoidable jargon of the field is explained in a glossary, and the narrative constructs a methodical and transparent overview of how Welsh influenced Tolkien’s story line, as well as his synthetic languages Quenya and Sindarin.

The overview is based on specific examples of attested names, placed in the context of their linguistic and cultural background, while highlighting the peculiar features of Welsh, the “the senior language of the men of Britain” (MC 189), that Tolkien found so intriguing.

Examining Tolkien’s linguistic creations and Legendarium specifically through the lens of Welsh produces a “myopic” vision of his work, but intentionally so, because as Jane Chance said in an interview, “the northern European influence seems more important than the Celtic, from what I have been able to tell. Perhaps that is because so much of the work done on Tolkien’s medievalism thus far has focused on the northern European influence.”[1] This book, therefore, takes a step outside the northern-European box around Tolkien Studies to show that there is more to it than that.

This project has been in preparation for some time, and I was at first concerned when the announcement of Carl Phelpstead’s Tolkien and Wales came to my attention, but was relieved when I got my copy, and saw that his excellent book had not touched on any of my topics. The reason for this is found in Chapter 3 (“Inventing Language”) of Phelpstead’s book, where he says that Tolkienian Linguistics is “a specialized discipline” that he considers “too technical” for his intended audience. (p. 39) Many of my topics, in other words, fall outside what Phelpstead considered his purview. It might, therefore, be said that Tolkien and Welsh fills in the linguistic gap in Tolkien and Wales.

Though some of my research would have fit in Phelpstead’s other chapters, it had not been previously published, and was, therefore, unavailable to him. There is, therefore, something for non-linguists in Tolkien and Welsh as well.

Tolkien explained that Sindarin was intentionally “devised to give it a linguistic character very like (though not identical with) British-Welsh: because that character is one that I find, in some linguistic moods, very attractive; and because it seems to fit the rather ‘Celtic’ type of legends and stories told of its speakers.” (L.176)

In “English and Welsh” (MC 197 n.33), Tolkien commented that the names and places in The Lord of the Rings were primarily developed “on patterns deliberately modeled” on Welsh sources, but not identical with them. The Welsh components of his tale, concludes Tolkien, are what have “given perhaps more pleasure to more readers than anything else in it.”

While there are those who might not agree with Tolkien’s assertion, I am not one of them. In his book on Welsh folklore,[2] Sikes remarks that although Keightley[3] took Shakespeare to task in his Fairy Mythology for the inaccuracy of his use of “English fairy superstitions,” no such thing could be said of the Bard’s use of Welsh folklore. Shakespeare’s knowledge and use of Welsh fairy motifs and lore, notes Sikes, were “extensive and peculiarly faithful.” The same can be said of Tolkien.

Tolkien was himself well aware of Shakespeare’s use of Welsh, and commented on it in a letter. (L.320)

In another letter, Tolkien commented that his Legendarium was “fundamentally linguistic in inspiration.” In Tolkien’s creative process, a name came first and the story of the name followed. His literary world was created to provide a place where his names could be at home. (L.219) Tolkien also noted that he liked history and was moved by it, especially such history as “throws light on words and names.” (L.264)

Tolkien further explained that the process behind the creation of Middle-earth was an idiosyncratic enterprise undertaken to satisfy his own private linguistic taste. He was, therefore, not surprised that most analyses of his work went awry because “linguistic invention” is a “comparatively rare” art form, and most analysts have little understanding “of how a philologist would go about it.” Their analyses “appear to be unauthentic embroideries on my work,” said Tolkien, “throwing light only on the state of mind of [their] contrivers, not on me or on my actual intention and procedure.” (L.380)

What makes this book different from other books about Tolkien is that its author is a linguist who shares Tolkien’s appreciation of the histories of words and names, and who plays at the same kind of linguistic invention himself. It is a linguistic perspective that begins with a name or a word, and looks for its story in the real world with which Tolkien was familiar.

A particularly important source of material for the study of Tolkien’s use of Welsh are the books by Sir John Rhys, the Professor of Celtic Languages at Oxford when Tolkien was a student there. Companion and Guide: Chronology details the days of the week and times of day that Tolkien probably took classes on Welsh (The Mabinogion) from Rhys in 1914-1915. (pp. 50, 52, 55, 59)

As any assiduous student should know, when you take a course from someone who has written a book on the topic of the course, the book will be a part of the course, even if it is not on the required reading list. The probability, therefore, that Tolkien was aware of the material in Rhys’ books is very high. The parallels between the material in Rhys’ books and that in Tolkien’s Legendarium make it almost a certainty that Tolkien read Rhys.

Max Förster’s Celtic Vocabulary in English[4] is another significant work in Tolkien’s understanding of Welsh. Tolkien cited this seminal study of Celtic loanwords in Old English in “English and Welsh.” (MC 167) Förster is likewise cited in the notes of the translation of Sir Gawain and the Green Knight that Tolkein edited along with E.V. Gordon.[5] Förster (1869—1954) was a well-known German linguist specializing in English. He was the head of the Department of English Philology at Munich University from 1925 to 1934.

In discussing Förster’s book, Tolkien said “many ‘English’ surnames, ranging from the rarest to the most familiar, are linguistically derived from Welsh (or British), from place-names, patronymics, personal names, or nick-names; or are in part so derived, even when that origin is no longer obvious.” (MC 167) He went on to explain that simply having a Welsh name, however, did not make someone Welsh, any more than his own German surname made him a German. (MC 170) Tolkien wanted to temper the interest in surnames, “to which Welsh is often the key,” by recalling “the age-long interpenetration of the peoples speaking English and Welsh.” (MC 167)

It would have been impossible, therefore, for a man who selected the names in his Legendarium with care, to ignore the Welsh linguistic layer found in ‘English’ names, when he was creating his “parody” of the toponymy “of rural England” (L.250) It should, then, come as no surprise that ‘Welsh is often the key’ to the names of Middle-earth, especially the Hobbit names.

Even though Tolkien remarked on his failure to grasp Old Irish or its modern descendant (L.134), he was not unfamiliar with it. In a letter, he states that he knows many Celtic things in their original languages of Irish and Welsh. (L.26) Tolkien also preserved the ‘P-Celtic’ and ‘Q-Celtic’ distinction from Celtic linguistics in the Elvish languages. The distinction is based on the way that the Proto-Celtic *kw evolved in each of them. Sindarin and Welsh are ‘P’ languages, and Irish and Quenya are ‘Q’ languages. An Irish layer in Tolkien’s Celtic linguistic creations, therefore, should come as no surprise.

Rhys observes that based on the examples of the language spoken by the Fairies of Dyved provided by a man who “spent a portion of his youth underground among” them, the language of the Fairies resembled “Welsh with a touch of Irish and Greek.”[6] Tolkien’s Elvish languages match the description well enough. Welsh with a touch of Irish is more than evident. The Greek less so, but it is there, most prominently in the name Erebor (Q.V.), the mountain where Smaug horded his treasure trove.

Though Tolkien said that his nomenclature was modeled on Welsh sources, but not identical to them, there are occasionally words and names that are in fact identical. Most often, they are found in the early drafts of his manuscripts published as The History of Middle-earth (HoMe). When they are, the degree of certainty of the story told about the word is high. When they are not, we find ourselves in the realm of most linguistic studies of names. Reaney highlights this characteristic of linguistics in The Origin of English Place Names, lamenting the fact that “we are often concerned with possibilities or probabilities rather than with definite etymologies.”[7] That is part and parcel of this particular branch of study.

Language is full of ambiguities, which is what makes it such an interesting topic. The number of linguistic jests, puns, and bilingual diplosemes in Tolkien’s work suggest that he never met an ambiguity that he didn’t like. The ambiguities, therefore, are a part of this study, and are often the key to understanding the story behind the name.

This volume applies a semantic analysis to Tolkien’s invented nomenclature, comparing it with the Celtic languages, but in particular with Welsh, to determine the likely first-world history and context of the Celtic roots that Tolkien used. The focus is on sources that were current at the time in which Tolkien lived and wrote. Modern theories may have supplanted the theories of Tolkien’s time, but that is irrelevant. This volume explores the question of what Tolkien thought, not what we think we know now.

Linguistic invention is a product of the mind, and is, therefore, not governed by the fixed and immutable rules of the sciences. As Albert Einstein once said, imagination is greater than knowledge. Since Tolkien is no longer with us, no “proof”—in the scientific sense—is possible. The facts, therefore, must speak for themselves when no conclusion is possible.

This book combines both previously published and unpublished essays to bring together all the author’s work on this topic in one convenient volume. Many of the previously published essays have been especially revised and expanded.


[2] Wirt Sikes, British Goblins: Welsh Folk Lore, Fairy Mythology, Legends and Traditions, Boston: James R. Osgood & Co., 1881, p. 14.

[3] The reference is to: Thomas Keightley, The Fairy Mythology: Illustrative of the Romance and Superstition of Various Countries, London: H. G. Bohn, 1850.

[4] Max Förster, Keltisches Wortgut im Englishen: Eine Sprachlige Untersuchung, Halle: Verlag van Max Niemeyer, 1921.

[5] Sir Gawain and the Green Knight, J.R.R. Tolkien, Eric Valentine Gordon (eds.), The Clarendon Press, 1925, p. 115.

[6] John Rhys, Studies in the Arthurian Legend, Clarendon Press, 1891, p. 74n.

[7] P.H. Reaney, The Origin of English Place Names, London: Routledge and Kegan Paul, 1960, p, 72.

To learn more about the book, follow the links below.

copyright bar