Oxford University Press's
Academic Insights for the Thinking World

A definition of ‘hobbit’ for the OED

In a hole in the ground there lived a hobbit… What’s a hobbit and how did J.R.R. Tolkien come by this word? Was it invented, adapted, or stolen? To celebrate the release of The Hobbit film and renewed interest in J.R.R Tolkien’s work, we’ve excerpted this passage from The Ring of Words: Tolkien and the Oxford English Dictionary by Peter Gilliver, Jeremy Marshall, and Edmund Weiner.

The OED entry for HOBBIT was published in 1976, in Volume II (H-N) of the Supplement to the OED. In December 1969, when this entry was in preparation, the Editor, R. W. Burchfield, sent a draft of the entry to Tolkien (who had been his teacher), asking for his comments. Tolkien did not reply until the following August: 1970 had turned out to be a year of ‘great pressures for me and domestic distresses’, as he explained in an apologetic letter to Burchfeld. Fortunately there was still time to make use of his comments, which he included in a further letter (cf. Lett. 316) dated 11 September 1970:

Unfortunately, as all lexicographers know, ‘don’t look into things, unless you are looking for trouble: they nearly always turn out to [be] less simple than you thought’. You will shortly be receiving a long letter on hobbit and related matters, of which, even if it is in time, only a small part may be useful or interesting to you.

For the moment this is held up, because I am having the matter of the etymology: ‘invented by J.R.R. Tolkien’: investigated by experts. I knew that the claim was not clear, but I had not troubled to look into it, until faced by the inclusion of hobbit in the Supplement.

In the meanwhile I submit for your consideration the following definition:

One of an imaginary people, a small variety of the human race, that gave themselves this name (meaning ‘hole-dweller’) but were called by others halflings, since they were half the height of normal Men.

This assumes that the etymology can stand. If not it may be necessary to modify it: e.g. by substituting after ‘race’

; in the tales of J.R.R. Tolkien said to have given themselves this name, though others called them…..

If it stands, as I think it will even if an alleged older story called ‘The Hobbit’ can be traced, then the ‘(meaning ‘hole-dweller’)’ could be transferred to the etymology.

This definition, since it is more than twice as long as the one that you submitted and differs from it widely, will need some justification. I will supply it.

The ‘long letter on hobbit and related matters’, in which presumably the ‘justification’ was going to be supplied, was never received, and no more was heard of the ‘investigation by experts’.

The entry which was actually published and now appears (with the addition of Tolkien’s death date) in the OED (Second Edition) reads:

In the tales of J.R.R. Tolkien (1892-1973): one of an imaginary people, a small variety of the human race, that gave themselves this name (meaning ‘hole-dweller’) but were called by others halflings, since they were half the height of normal men.


Are not these the Halflings, that some among us call the Holbytlan? (LR III. viii)

Having as far as he knew invented the word, Tolkien provided an imaginary etymology for hobbit, in order to fit the word into the linguistic landscape of Middle-earth. This was a remarkable feat of reverse engineering, not quite like any of his other etymological exploits amongst the tongues of Middle-earth.

On encountering the Rohirrim, the hobbits notice that their speech contains many words that sound like Shire words but have a more archaic form. The prime example is their word for the hobbits themselves: holbytla. This is a well-formed Old English compound (because Tolkien represents the language of the Rohirrim as Old English). It is made up of hol ‘hole’ and bytla ‘builder’; it just happens, as far as we know, never to have existed in Old English, and if hobbit turned out to be a genuine word from folklore it is most unlikely that this would be its actual etymology.

Tolkien is playfully suggesting that if there had been an Old English word holbytla (its accentuation would have been similar to that of hole-builder) it might well have come down into modern English as hobbit (though actually a more likely form would be hobittle, with a first syllable like that of hobo). But just to complicate things further, in a note at the end of the Appendices to The Lord of the Rings, Tolkien is at pains to explain that ‘really’ the human languages of Middle-earth were quite different from the English and Old English which he has ‘translated’ them into. Hobbit was ‘really’ kuduk and holbytla was ‘really’ kûd-dûkan; but he claims to have devised the English equivalents in order the better to convey the flavour of the world he is writing about.


The first quotation in the OED entry is the famous first line of The Hobbit: ‘In a hole in the ground there lived a hobbit.’ It is well known that Tolkien scribbled this sentence on an examination paper he was marking in a moment of boredom (Lett. 163). The word is Tolkien’s most famous coinage–if it is indeed a coinage. His letter of September 1970 to the Dictionary department (quoted above) makes it clear that Tolkien was not entirely certain that he had invented the word, but neither he nor anyone else had at this time uncovered any earlier instance. A correspondent to The Observer on 16 January 1938 had claimed to recall the word from an earlier story, but this could not be traced; Tolkien’s response (Lett. 25) shows that he at least acknowledged the possibility, though he commented: ‘I suspect that the two hobbits are accidental homophones.’ In another letter written in 1971 on the same subject (Lett. 319), he wrote of his ‘unsupported assertion that I remember the occasion of its invention…. However, one cannot exclude the possibility that buried childhood memories might suddenly rise to the surface long after.’

Then, after Tolkien’s death, an example of the word did turn up, in a long list of ‘supernatural beings’ appearing in the so-called Denham Tracts, compiled by the Yorkshire merchant M. A. Denham (1800 or 1801-1859). Denham was an amateur folklorist who published many books and pamphlets, including twenty Minor Tracts on Folklore (1849-c.1854). The majority of these Tracts were collected in an edition prepared for the Folklore Society in the 1890s, and the word hobbit appears in the second volume (1895) of this edition.

The discovery of the word in the Denham Tracts was reported in The Times on 31 May 1977. The article records that Tolkien, when asked whence he had got the name, ‘replied that he could not remember: perhaps he invented it; or “I may have picked it up from a nineteenth century source”.’ (Perhaps Tolkien still recalled that exchange of letters in 1938.) The Times writer rather boldly asserted that this ‘nineteenth century source’ had now been identified as Tolkien’s inspiration. But could Tolkien have read the relevant Denham Tract? It certainly seems an unlikely origin for ‘buried childhood memories’.

The Denham Tracts are a bibliographically untidy collection of texts. It seems that Denham would write a short piece and have it printed in a relatively small number of copies which were then distributed, probably mostly to his own circle. There is no single complete collection in existence, and in any case this would be hard to achieve since it also seems that Denham used to publish rewritten versions of his Tracts.

We know of four versions of the Tract in question. The original version (to which the others refer back) is in fact an article entitled ‘Seasonable Information’ published in the Literary Gazette of 23 December 1848 (p. 849, column 2). It is really no more than a light-hearted joke for the Christmas season, along the following lines: the glossarist and folklorist Francis Grose observes (says Denham) that those born on Christmas Day cannot see spirits; what a happiness this must have been seventy or eighty years ago for those who had the luck to be born on this day, when the whole earth was overrun with them–and there then follows a list of 131 ‘supernatural beings’ of various kinds, beginning with ‘ghosts, boggles, bloody-bones, spirits, demons’, and ending with ‘silkies, cauld-lads, death-hearses, and goblins and apparitions of every shape and make, kind and description’. The list contains beings drawn from both classical tradition and British folklore. Some are very familiar, some are evidently drawn from Denham’s own research, and some appear to be unique to the list. His sources for the lesser-known words are uncertain, but he does share material with such sources as Reginald Scot’s Discoverie of Witchcraft (1584). It is not a scholarly piece but just an opportunity to exhibit the vastness of the vocabulary of supernatural beings.

The second version (1851) is a separate Tract with a title page inscribed: ‘To all and singular the Ghosts, Hobgoblins, and Phantasms, of the United Kingdom of Great Britain and Ireland, These brief Pages are Fearlessly Inscribed, In utter defiance of their Power and Influence, By their verie hvmble Seruaunte, To Com’aund, M:A:D.’ The list is essentially the same but now contains 165 items. The new names have been added en bloc at the end of the list, with a few rearrangements and changes in spelling, and seven quasi-scholarly footnotes. The third version (1853) is similar, with almost the same title page. It contains 198 items and 31 footnotes. The additional new names have been inserted almost, but not quite, at the end of the list, i.e. within the section that was inserted in 1851. The word hobbits is the first word of the new section, which continues with ‘hobgoblins [a repetition: the word is already in the original list], brown-men, cowies, dunnies, wirrikows…’. Unfortunately ‘hobbits’ is not footnoted. The fourth version is the reprint in the Publications of the Folk-lore Society mentioned above, where the word hobbits was first detected. This is in all important respects identical to the 1853 version. A very small number of changes appear to have been made, perhaps by the editor responsible for the republication.

The 1895 version would have been readily available in university libraries accessible to Tolkien (there is a copy in Oxford), and he was interested in folklore. Alternatively, he could have seen a list copied out by one of his friends or colleagues — someone like C. S. Lewis who read all kinds of abstruse writings. If there were any other unusual items in the list which also occurred in Tolkien’s writings, we might suspect that the Tract was the source for all of them; but even though such curious words might have been quite handy for some of his more light-hearted poems, there is no trace of them. There seems to be nothing that tips the scales in favour of the theory that he had somehow come across the word from the Tract.

Even if Tolkien had in fact picked the word up from the Tract, this would only replace one mystery with another, for we do not know where Denham found the word, or what its meaning and etymology are. The other words newly added to Denham’s list do not seem to be traceable to particular sources that might contain hobbit, unnoticed. It is possible that the first syllable hob– is the same as that of the next word in the list, hobgoblin. This element hob (OED: HOB n. 1, sense 2a) has the same meaning as the full word hobgoblin (but is about a century or so older). It originated as a familiar form of Rob, short for Robert (of which Robin, used in the sprite name Robin Goodfellow, is a diminutive). It could be hypothesized that hobbit is a derivative of this word hob; the ending –it could be explained as the diminutive suffix more usually spelt –et, found in midget, moppet, and snippet, and the word would mean ‘a small goblin or sprite’.

Another idea is that it might be a shortening of hobbity-hoy, a variant (quoted by the OED from a glossary of Yorkshire dialect) of hobbledehoy, meaning ‘a clumsy or awkward youth’: of course this is not a supernatural being, but not all Denham’s words actually are. Either of these suggestions requires the assumption that it was a genuine word used by country people and recorded by Denham or an informant of his, but by no one else.

Alternatively, could the word have been suggested to Tolkien by something other than the Tract? There is the distinct possibility that a similar word (not necessarily with the meaning he gave to it) listed in the OED might have lodged, submerged, in his memory. One candidate is the word hobbity-hoy already mentioned. Another is the entry howitz, haubitz, meaning ‘howitzer’, which has the variant spellings (unchanged in the plural) hobbits or hobits, illustrated by the following quotations:

Hobits are a sort of small Mortars from 6 to 8 Inches Diameter. Their Carriages are like those of Guns, only much shorter.
(J. Harris Lexicon Technicum II (1710))

Little Hobbits charged with the various kinds of Fire-Balls.
(G. Shelvocke Artillery (1729) V. 377)

When one recalls that Bilbo meant a kind of sword and was often used in the 17th and 18th centuries as the name of a sword personified, the idea seems quite attractive. However, this is just one more unproven speculation; the real origin of Tolkien’s word remains obscure.

Recent Comments

There are currently no comments.