Monday, 13 July 2020

Great-Granduncle Bullroarer

This passage in chapter 1 of The Hobbit is very familiar:

If you have ever seen a dragon in a pinch, you will realize that this was only poetical exaggeration applied to any hobbit, even to Old Took’s great-grand-uncle Bullroarer, who was so huge (for a hobbit) that he could ride a horse. He charged the ranks of the goblins of Mount Gram in the Battle of the Green Fields, and knocked their king Golfimbul’s head clean off with a wooden club. It sailed a hundred yards through the air and went down a rabbit-hole, and in this way the battle was won and the game of Golf invented at the same moment.

But why did Tolkien name the Old Took’s great-granduncle ‘Bullroarer’?

Tolkien himself said (Nomenclature, s.v. Took):

I believed when I wrote it that bullroarer was a word used by anthropologists, etc. for an instrument(s) used by uncivilized peoples that made a roaring sound.

And, he continued:

But I cannot find it in any dictionaries (not even in O.E.D. Suppl. [i.e. the Supplement]).

There is more to be said, however, about the word, the thing, and Tolkien’s use of it. Arden R. Smith (2000) has already written about this; unfortunately I have been unable to consult the article, but I believe he points to the discussions of the bullroarer in Fraser’s Golden Bough and (more significantly) the writings of Andrew Lang.

1. The word bullroarer

As was pointed out by Hammond and Scull (2005: 6), among others, bullroarer does appear in the OED under the entry for BULL n.1; in the first edition, available to Tolkien, it is listed, with an 1881 quotation, but not defined; in the Supplement it has only half a definition, which merely states ‘spec. used by Australian Aborigines in certain (religious) ceremonies’. It is certainly in the context of Aboriginal life, or at least of the anthropological study of traditional peoples, that the word is probably familiar to many people nowadays, just as Tolkien said it was to him. OED’s three quotations support this, in all evidently referring to this kind of setting. The first quotation, however, points to earlier use in a domestic British context and explains what a bullroarer is:

1881   Academy 9 Apr. 263/3   A flat slip of wood a few inches long, narrowing to one or both ends, and fastened by one end to a thong for whirling it round, when it gives an intermittent whirring or roaring noise, heard a long way is known as a country boy’s plaything in Europe, called in England a ‘whizzer’ or ‘bull-roarer’.

This comes from an article written by Edward Burnett Tylor (1832–1917), a leading British anthropologist, the first Reader in Anthropology at the University of Oxford, who (surely no coincidence) was the anthropological consultant to the first edition of the OED. In an address to a special session of the the Anthropological Society of Washington on 11 October, 1884 (Transactions of the Anthropological Society of Washington 3), Tylor speaks of some exhibits in an ethnological museum:

Certain curious instruments..consisted simply of flat, oblong, or oval pieces of wood, fastened at the end of a thong, so as to be whirled round and round, causing a whirring or roaring noise. (1884:86)

These particular examples were from the Ute and Zuni peoples of North America, but Tylor observed that the same instrument was used by peoples in Africa and Australia, usually in religious ceremonies, and the point he went on to make was: is the worldwide distribution of the instrument due to transmission from common primeval ancestors or merely to independent invention — a question of development similar to the kind of questions in the realm of language investigated by the OED and philologists like Tolkien.

In the course of this discussion he goes on to say:

It is a toy well known to country-people, both in Germany and in England. Its English name is the ‘bull-roarer’; and, when the children play with it in the country villages, it is hardly possible (as I know from experience) to distinguish its sound from the bellowing of an angry bull. (1884:87)

Tylor’s 1881 article was a review of a pioneering study of Australian aboriginal beliefs, Kamilaroi and Kurnai (1880), published by  the anthropologist and explorer Lorimer Fison (1832–1907) and the Australian explorer Alfred William Howitt (1830–1908). In Appendix E of this work, Fison introduces the term bullroarer to anthropological discourse in his discussion of the Turndun (the name in a particular Aboriginal language for the instrument).  Howitt obtained several bullroarers from Aboriginal peoples and gave them to E. B. Tylor, who in turn passed them to the Pitt Rivers Museum in Oxford (see the Howitt-Tylor papers on the Pitt Rivers Museum website).

Fison too refers to the toy being used by English children, and names it bullroarer:

It much resembles in general character the wooden toy which I remember to have made as a boy, called a bull roarer. (1880:267)

He would have been making his bullroarer in around 1840. He was born in Barningham, Suffolk; as we shall see below, what evidence we have for the provenance of the word bullroarer comes from East Anglia.

In much the same way, the folklorist and writer Andrew Lang (1844–1912) begins his essay (1884) by relating the sacred instrument of the tribesperson to the schoolboy’s toy and again names it bullroarer.

As the belated traveller makes his way through the monotonous plains of Australia…he occasionally hears a singular sound… If he be an Englishman, country-bred, he says to himself, ‘Why, that is the bull-roarer’. If he knows the colony and the ways of the natives, he knows that the blacks are celebrating their tribal mysteries…The instrument which produces the sounds that warn women to remain far away is a toy familiar to English country lads. They call it the bull-roarer. 

There are a couple of slightly surprising aspects to all this. The first relates to the word itself. Despite all these anthropologists averring that bullroarer is its traditional and well-established name, there is very little on record that definitely confirms this, as far as I have been able to ascertain, in any text prior to Fison’s 1880 statement. None of the many dialect dictionaries produced during the nineteenth century seem to include the word; it is not found in Joseph Wright’s monumental English Dialect Dictionary, which one would expect to have picked it up. The earliest example of the word I have been able to find applies it to a broken-winded horse (often described simply as a roarer):

When the horse is led out if you suspect his wind, turn him short, and give him a blow with your hand or stick if he grunts, reject him. Such a horse is a bull roarer. (Devizes & Wiltshire Gazette  17 Dec. 1840, 4/4)
As regards later use, Arthur Ransome used the word to mean a nautical instrument, apparently wind-operated, for attracting the attention of other vessels:

Cap’n Flint Sir, would you mind now giving another three blasts on that bull-roarer? (Peter Duck (1932), chapter 13)

Apart from that, I have not been able to find any occurrences of the word outside the context of anthropology, ethnology, musicology, and similar learned subjects.

There is indeed one earlier occurrence which classes it among children’s ‘amusements’ that make a characteristic noise, but the context gives no clue as to whether the word refers to a piece of wood (or similar material) whirled round to produce a roaring sound:

For explaining the theory of sound, the whistle, the humming-top, the whiz-gig, the pop-gun, the bull-roarer, and sundry other amusements, well known in the playground, would amply suffice. (Illustrated London News 8 Nov. 1851, 19/2)
So closely connected with anthropology has the word become that anyone who speculates about its origin is quite likely to think vaguely, as I have, that it is some kind of loan-translation from the language of a traditional people: if not an Aboriginal one (seeing that they did not have bulls), perhaps a Native American or African one. It just sounds foreign. 

There are, as we shall see, a few further bits of evidence for the word’s use in England from non-literary sources, but still none certainly older than 1880.

2. The thing itself

The instrument is known all over the world, and has held an important place in the rituals of many cultures. It has also been known for a very long time: in ancient Greek it was known as rhombos (as the OED tells us in the etymology of RHOMBUS n.), and because the Greek version was diamond-shaped, the name was transferred to the geometrical figure, as well as to a species of fish.  

But there is a second mildly surprising aspect to this. While Fison, Tylor, Lang, and other writers from 1880 onwards assure us that British children played with this toy all the time, there seems to be very little corroboration from other sources. The article in the Illustrated London News cited above may support their assertion, if the word bull-roarer there does denote this particular toy. Another article, albeit later in date than the various references from our quartet of anthropologists, probably does support it, though again, it gives no description, assuming that the reader knows what a bullroarer is:

Nobody who loves a quiet life will give boys drums, trumpets, fireworks, hammers, knives with an unheard of number of blades, nor the toy known as the ‘bull roarer’, nor pea-shooters, squirts, [etc.]. (Daily News 24 Dec. 1887, 5/2)

It is noticeable that this speaks of giving a boy a bullroarer, whereas the anthropologists imply that boys were in the habit of making them for themselves.

There is also a certain amount of linguistic evidence to support the anthropologists’ assertion, in the form of synonyms for bullroarer, though few of them are older than 1880. The OED gives thunderer (1908), whizzing-blade (1905), whizzing-stick (1890), whizzer (1881, as we have seen above), humbuzz (before 1878), and perhaps most promisingly, the Scottish bummer (OED BUMMER n.2 1a), which is helpfully defined in the first example (1821) as ‘a thin piece of wood swung round by a cord’. There is also the much older hurr or hurr-bone (1483), but as this is glossed by the Latin giraculum it may refer to another child’s toy, the whirligig, a disk pierced by two holes near the centre with looped strings passing through them, which is made to rotate by winding up the strings and pulling their looped ends apart.

A supportive body of material comes from another leading British anthropologist, Alfred Cort Haddon (1855–1940), of the University of Cambridge, who was an expert on New Guinea and its islands. Chapter 10 of his book The Study of Man is entitled ‘the Bull-roarer’. He writes:

For some years past I have collected all the specimens and information I could about this interesting object.

His collection came not from traditional societies overseas but from the British Isles. He gives a run-through of all his relevant materials, consisting partly of his own specimens (with their shapes and dimensions, and accompanied by a page of illustrations), and partly of information from other unspecified sources. This may be best summarized in a list of quoted extracts (omitting the shapes and dimensions, which, though interesting, are not relevant to the present enquiry) (1898: 219–224):

1. ‘Specimen made by a boy at Balham in Surrey (London, S.W.)’ (term for item not given)
2. ‘I have heard of it in Essex, but have not seen a specimen’
3. ‘In West Suffolk it is called a “hummer”’
4. ‘I have several specimens from different parts of Norfolk, where it is called “humming buzzer”, or simply “buzzer”’
5. ‘One specimen “buzz” from Mid-Norfolk’
6. ‘I have been informed that in Cambridgeshire it was called a “bull”
7. ‘In Bedfordshire its name is “buzzer”’
8. ‘The Lincolnshire variety, “swish”’
9. ‘I have heard of its occurrence in the East Riding of Yorkshire, but have no details’ 
10. ‘In East Derbyshire it is known as a “bummer” or “buzzer”’
11. ‘A model of a Warwickshire type…’
12. ‘Another model, also called “bummer”, said to be used in Warwickshire, Staffordshire, and Shropshire’ (Of these last two he writes ‘I must confess that I am not satisfied about these last two implements. I have one or two others that were given me by the same friend which vary considerably in form, and had no localities given with them.’)
13. ‘The Rev. Elias Owen, of Oswestry, kindly had a “roarer” made for me as they were used sixty years ago in Montgomeryshire in Wales’
14. ‘I have been told that the bull-roarer was known as a “thunder-spell” in some parts of Scotland, and in Aberdeen as a “thunder-bolt”’ (He adds a footnote, part of which states: ‘Mr. W. S. Laverock, of the Liverpool Museum, has informed me that “thunner spells” are quite common in Aberdeenshire and Kincardineshire; they were made by farm-servants and villagers’)

Finally, he says:

15. I have only two notices of the bull-roarer from Ireland — one from the town and county of Cork, the other from Ballycastle, County Antrim, where the Rev. J. P. Barnes kindly gave me a specimen… Its use is very local, but I am informed that the schoolboys in Coleraine often make them. Mr. Barnes writes: ‘From inquiry made, I come to the conclusion that the “Bull-roarer” (its local name) is not indigenous, but an importation. The boy who gave me this says he got the idea from his father, who is a coastguard; his father once tied a string to a piece of wood lying near the fireside, and began to twirl it round for the children’s amusement, saying, “That’s what I have seen n—ers [my suppression] do in the West Indies”.’

And in another footnote he adds: 

16. ‘I have been informed that “boomers” are in common use among boys in County Down.’

We are indebted to Haddon for collecting a set of terms which the dialectologists seem to have entirely overlooked. It is noteworthy, however, that he never directly attributes the name bullroarer to any district in England, though he locates bull in Cambridgeshire and roarer in Wales. He at first appears positively to assert that bullroarer is the local name in Coleraine, but then goes on to quote J. P. Barnes to the effect that the toy may have been introduced from the West Indies, implying in turn that the word may have come with it. The indication that the name is boomer in County Down may also be a pointer to the fact that bullroarer was not an indigenous term in Northern Ireland.

A further body of evidence comes from the catalogue of the collections of the Pitt Rivers Museum in Oxford. This excellent database lists all 254 bullroarers in the museum’s collections, including full details of the shape and size, place of origin, maker, name of the collector, date of collection, and so on, and usually also a photograph. Round about 25 of these come from the United Kingdom (one or two seem to be recent artefacts, deliberately made, and can therefore be excluded from consideration). Of particular interest is the fact that the local name is sometimes given. The following table presents the origin, local name, and date of collection of the older ones.

Local name
When collected
whizzing stick, buzzer
< 1902
< 1905
Tewkesbury, Glos.
whizzing stick, buzzer
< 1902
Needingworth, Cambs.
roaring bull
< 1902
St Ives, Cambs
bull roarer
< 1902
St Ives, Cambs
< 1902
Bungay, Suffolk
< 1894
< 1913
< 1902
< 1904
< 1904
Clanfield, Oxfordshire
< 1920
< 1939
< 1939
< 1902
Ashdown Forest, E. Sussex
Chichester, W. Sussex
whizzing blade
Chichester, W. Sussex
whizzing blade
Ballycastle, Co. Antrim
Co. Antrim
< 1939
srannan [from Gaelic]
< 1917
Scotland Elgin Grampian Knockando Parish
thunner-spehl [sic]
Scotland Aberdeenshire [Grampian] Pitsligo
Duxford, Cambs.

In comparing this list with Haddon’s we need to be cautious, as it’s not impossible that the collectors or curators could have consulted Haddon himself or his book when assigning names to the artefacts. If the two sets of names are independent, it is certainly noteworthy that the following agreements occur:

H. 7 and P-R. 1: Bedford, buzzer
H. 4 and P.-R. 8: Norfolk, buzzer
H. 14 and P.-R. 23: Aberdeenshire, thunner-spell

Also, that each list records hummer but from widely separated places: H. 3 West Suffolk, P.-R. 16  East Sussex.

Most important, of course, is the occurrence of bullroarer (P.-R. 5) from St. Ives, Cambridgeshire, if this is not simply the application of the established anthropological name rather than a genuine record of the local one. Its authenticity is supported by the fact that the collector, Miss F. E. Ogden, also collected P.-R. Nos. 1, 3, 5, and 6, for each of which she seems to have been careful to record the local name, and also by the occurrence of the similar roaring bull at Needingworth, Cambridgeshire (P.-R. 4) and  bull in Cambridgeshire (H. 6).

Perhaps we can assume therefore that the toy was in use for the best part of the nineteenth century (at least) in Britain and Ireland. We can also surmise that it was only called bullroarer in a very limited area around Cambridge, St Ives, and Needingworth, perhaps extending as far as Barningham, Suffolk, just over 50 miles to the east, when Lorimer Fison, a native of the district, introduced it into anthropological literature in 1880.

Another slight puzzle remains in the fact that few if any, descriptions of, or allusions to, its use, exist in literature. It seems surprising that it should never have been written about, since a small boy swinging a piece of wood rapidly round his head on a longish piece of string is a serious inconvenience, and indeed a danger, while the noise it makes soon loses its charms, as Andrew Lang notes:

The common bull-roarer is an inexpensive toy which anyone can make. I do not, however, recommend it to families, for two reasons. In the first place, it produces a most horrible and unexampled din, which endears it to the very young, but renders it detested to persons of mature age. In the second place, the character of the toy is such that it will almost infallibly break all that is fragile in the house where it is used, and will probably put out the eyes of some of the inhabitants. (1884)

3. The Anthropologists

It is worth noting that Fison and Howitt, Tylor, and Lang were all closely in touch with one another. Andrew Lang, as a literary folklorist, was probably dependent for information on the others, who were of course field researchers. An article from 1896 in the comic magazine Judy, which was a less successful and shorter-lived counterpart of Punch, suggests  that Lang’s interest in the bullroarer was a matter of widespread knowledge and a source of humour. The column ‘Judy’s Literary Kettledrum’, under a subtitle ‘Guest of the Hour: Mr A—w L—g’, describes a visit from Andrew Lang to the magazine’s offices and a bantering conversation between him and ‘Judy’, the magazine’s equivalent of Mr. Punch. The column opens with a sardonic view of the anthropologist’s interpretation of items of personal equipment:

Leaning on his own inimitable style, which serves him alike for rapier, broadsword, golf-club, shield, walking-stick, umbrella, and, upon occasions, even as fishing-rod and scourge, our Guest entered the office last Thursday afternoon.

The conversation begins with a discussion of a review of a work (not named) by Lang in the ‘P.M.G.’, the Pall Mall Gazette; reflecting on the reviewer’s treatment of him, Lang remarks ‘I’d no idea I’d got in so many of my old chestnuts’, before moving on to a brief joking discussion of Lang’s distinction between custom and myth. Then the visit ends in the following way:

‘Perhaps not,’ said the Prince of Parodists [i.e. Lang], rising to take leave. ‘Hark!’ he cried, as he turned towards the door, ‘surely that is the Bull-Roarer!’ And he hastened out to investigate. But the Staff smiled, for they knew that it was only the voice of Wilfred, whom the Business Manager had overtaken in a fault.

This seems to establish not just Andrew Lang’s reputation for having various idees fixes, but the fact that the bullroarer was a particular obsession of his, regarded as a source of amusement by others.

Indeed Lang, in his essay (1884), states ‘To study the bull-roarer is to take a lesson in folk-lore’. Alan Dundes, in the introductory part of his psychological study of the bullroarer (1976), details the anthropologists’ fascination with the bullroarer from the time when its use among Aboriginal groups was first described in the early 1880s down to the time of his essay. As regards the name, he confirms that it entered the formal anthropological lexicon in 1880 in Fison and Howitt’s Kamilaroi and Kurnai (as noted above), and that its widespread acceptance was furthered by E. B. Tylor in his review of the book in The Academy quoted by the OED (1881), and then by Andrew Lang’s essay (1884).  

Dundes explains that the contrast between the sacred object and the child’s plaything ‘was understood as exemplifying unilinear cultural evolution by means of which peoples progressed from savagery to civilisation having no visible record of such change other than a few tangible vestigial remains of the earlier period, remains which were dubbed survivals’ (1976:221). He also tells us that ‘since 1880 there has been a considerable body of scholarship devoted to the bullroarer’ (1976:221).

It is unnecessary here to describe the various theories attached to the bullroarer, covered by Dundes in some detail, and culminating in his own. For present purposes it is enough to highlight one point made by Dundes:

Part of the initial great interest in bullroarers had to do with whether or not they provided an indication of the existence of so called high gods among ‘primitive’ peoples [...] The sound of the bullroarer was likened to the sound of thunder which was thought to be the voice of a deity. (1976:223). 

One of those who pursued this line of thought was Robert Ranulph Marett (1866–1943), a British ethnologist, an exponent of the British Evolutionary School of Cultural Anthropology. From 1910 to 1936 he was Tylor’s successor as reader in social anthropology at the University of Oxford. He differed from Tylor and Frazer, the leading anthropologists of the time; he postulated an impersonal primitive religion, or ‘animatism’ (in contrast to Tylor’s ‘animism’), based on ‘awe’, a feeling of ‘submissiveness tempered with admiration, hopefulness, and even love’.

Marett devoted a whole essay in The Threshold of Religion (1910; reissued 1914) to the bullroarer. His subject was the existence of belief in a Supreme Being among ‘savage’ peoples, and his contention was:

A special group of these Supreme Beings, associated with the initiation rites of South-East Australia, has likewise in large part evolved out of a personification of the bull-roarer. (1914:145) 

He points out that he had already put this idea forward in an address to the British Association in 1899: this address is republished in The Threshold of Religion:

I have to confess to the opinion with regard to Daramulun, Mungan-ngaua, Tundun, and Baiamai, those divinities whom the Kurnai, Murrings, Kamilaroi, and other Australian groups address severally as ‘Our Father’, recognizing in them the supernatural headmen and lawgivers of their respective tribes, that their prototype is nothing more or less than that well-known material and inanimate object, the bull-roarer. (1914:16)

Just like Fison, Tylor, and Lang before him (though of course he may be at least partially drawing on them), he draws attention to the existence of the bullroarer as an English boy’s toy:

Natives of the Islands, if country-bred, may have had the opportunity in boyhood of cultivating a practical acquaintance with the bull-roarer under this, or some other, local designation of the toy, such as ‘roarer’, or ‘bull’, or ‘boomer’, or ‘buzzer, or ‘whizzer’, or ‘swish’ — names one and all eloquently expressive of its function. That function is, of course, to make a noise, the peculiar quality of which is best described by some such epithet as ‘unearthly’. (1924:155)

4. Marett and Tolkien

Robert Ranulph Marett was a classicist, but began moving towards anthropology with the publication of his paper on Pre-animistic Religion in Folklore (1890). The next year he became a Fellow of Exeter College, and remained as a tutor there until 1928, when he became Rector in succession to Lewis Farnell. He became Sub-Rector in 1893 and in the same year won the Green moral philosophy prize with ‘The Ethics of Savage Races’, which started his anthropology career. Although he challenged Tylor’s theory of animism, he was a strong supporter of Tylor, assisted him in the building up of social anthropology at Oxford, and became Reader in Social Anthropology in 1910 on Tylor’s retirement.

Tolkien and Marett knew each other, since Marett was Tolkien’s moral tutor, which means he did not teach him but was guardian of his welfare. They must at least have met to discuss Tolkien’s well-documented infringements of university discipline. But they also met at College Societies. Scull and Hammond (2006:56) document one occasion when Tolkien, as an undergraduate, was associated with Marett:

27 October 1914 Tolkien is very active at a meeting of the Stapledon Society, proposing a vote of censure, reporting a talk he had with the Sub-Rector concerning entertainment, and giving a warning to prospective officers. The Rector and Dr Marett lead a discussion of ‘Superman and International Law’ to which Tolkien also contributes.

Additionally, Marett was the founder of the Exeter College Dialectical Society, of which Tolkien was a member (Fry 2007:489). On 3 February 1914, Marett spoke there about ‘Origin and Validity in Religion’ and, at the other end of the Great War, on 28 October 1919, he gave his celebrated lecture ‘Bull-roarers and High Gods’ (I am indebted for this information to John Garth). Tolkien was certainly available to have attended both these meetings, the later one coming some time after he and his family had settled into their Alfred Street home.

On his return to Oxford from Leeds Tolkien could not have failed to encounter Marett, who continued in harness till his death during the Second World War. But was he acquainted with Marett’s publications? It seems likely. Though largely forgotten today, Marett was acclaimed in his lifetime. The journal Oceania described him in an obituary as ‘one of the world’s greatest ethnologists, a pioneer of Comparative Religion’ (1944). More to the point, on 25 October 1929 Marett delivered the Andrew Lang Lecture at the University of St Andrews, just as Tolkien was to do ten years later. It’s hard to imagine that Tolkien in 1939 would not have looked through the previous decade’s series of Lang Lectures to get an idea of what had already been said; nor that he would have omitted to see what his fellow Exonian in particular had had to say.

Marett’s Andrew Lang Lecture is entitled ‘The Raw Material of Religion’. He begins by paying homage to Lang:

I must count Andrew Lang as among my chief instructors… (1929:3)

Almost immediately the bullroarer, as a stock in trade of anthropology, makes an appearance:

I had my first taste of anthropology…when at Balliol in the middle eighties, my tutor, Strachan Davidson, advised me to read [Lang’s] Custom and Myth…assuredly it was for those days a sign of unusual broadmindedness in a student of classical antiquity thus to impress on the tiro, by vicariously strewing totems and bull-roarers in his path, the truth that history is as wide as the world of Man. (1929:3)

The rest of the lecture is based on his correspondence with Lang over many years. He has ‘a drawer-full of Lang’s letters’ ‘These dusty sheets retain a freshness which I reckon it a public service to try to waft abroad’ (1929:4)

A reason why the lecture might have engaged Tolkien’s interest, though not entirely his sympathy, soon appears: 

Among anthropological problems of lasting appeal the first in importance is surely that of the essential nature and value of human religion. (1929:5)

The topic of bullroarers soon returns:

Thus Baiame [an Australian ‘High God’] has such a duplicate and understudy in Tundun, a name said to mean ‘bull-roarer’. It was, in fact, largely on the strength of this etymological hint that I proposed to Lang the theory — I worked it up into an essay a good deal later on — that all the High Gods of Australia, prototypes and ectypes alike, were originally bull-roarers — not makers, therefore, so much as, specifically, makers of rain. (1929:13)

And more on page 18.

Tolkien may also have been aware of Marett’s practices when lecturing on anthropology around the country, assuming that the following report describes only one occasion among many:

Black magic in both savage and civilised communities was discussed last night at the Hull University College by Dr R. R. Marett, F.B.A., M.A., rector of Exeter College, Oxford professor of social anthropology at Oxford University. Dr Marett dealt with magic in relation to religion in a talk that was enlivened by stories of first-hand experiences. He was, he said, speaking as an anthropologist and not theologian. He talked of ‘bull-roarers’, a curious instrument for summoning, among other things, thunder and rain… But first of all he produced one from his pocket and gave a demonstration. Whirling the wooden blade at the end of a piece of string, he conjured up a roar. (Daily Mail (Hull), 12 Nov. 1936, 12/5)

My point then is that Tolkien is likely to have been aware of the centrality of the bullroarer in contemporary anthropological discourse, originally, to be sure, from his extensive reading of Lang’s works; but I guess also very likely to have been reminded of it by the writings and activities of this senior fellow and rector of his former college, given his interest in religion and myth, and especially considering that Marett was a Lang Lecture predecessor.

5. Tolkien and the bullroarer

The Took ancestor was not wielding a bullroarer but a wooden club when he knocked off the head of Golfimbul. My guess is that the bullroarer had become, for many people including Tolkien, a meme of the anthropological attitude to myth and religion, and a faintly absurd one at that. The idea of the bullroarer metamorphosing into a High God, would, I think, have struck Tolkien as ridiculous. And I suggest that Lang’s and Marett’s enthusiastic advocacy of the bullroarer could have been in Tolkien’s mind when, circa 1930, he started telling the story that became The Hobbit

In his bullroarer essay, Marett repeats the well-known anthropological principle ‘it is the ritual that generates the myth, and not the myth the ritual’ (1914:150), which he attributes to Robertson Smith (1889:17f). The account of Bullroarer Took’s action — ‘Golfimbul’s head...sailed a hundred yards through the air and went down a rabbit-hole, and in this way the battle was won and the game of Golf invented at the same moment’ — could be taken to represent a gentle satirizing of this principle. The game of golf, regarded as a modern ritual, is explained by a somewhat grotesque origin-legend, the smiting of Golfimbul’s head into a rabbit hole. The nickname ‘Bullroarer’ accompanies it as a further symbol of the author’s amused contempt for this kind of scholarship. Perhaps his very great size, ‘so huge (for a hobbit) that he could ride a horse’ is a faint nod towards the mythic size and power of the Ancestor in anthropological studies. But certainly there seems to be no other attribute of this member of the Took clan that could explain his possession of the nickname.

So much for the bullroarer and The Hobbit. But, as we know, all sorts of things lingered in the leaf-mould of Tolkien’s imagination. There are a few other suggestive sentences in Marett’s bullroarer essay that may be relevant to The Lord of the Rings.

On the terrifying effect of the sound of the bullroarer he says:

At the initiation ceremonies one of its functions is simply to frighten the uninitiated, and with that end in view it is personified as a terrible Hobgoblin by those who conduct the ceremonies. (1914:146)

Expanding on initiation ceremonies in south-eastern Australia, he says: 

The shuddering sound proceeding from the woods is explained to be the voice of Hobgoblin. No bloodless wraith is he, but an anthropomorphic being if ever there was one. (1914:158; italics mine)

This is part of a discussion upholding his belief (in opposition to the animism theory) that these supernatural beings are thought of as material: 

It might be worth while to inquire how far a universal source of anthropomorphic, as contrasted with animistic, that is, wraith-like, characters in supernatural beings is to be sought in personation. Thus there is reason to suspect that the manitu, whom the young American goes out into the woods to find, appears to him more often than not in the shape of a masked man. (1914:159; italics mine)

Now, I grant that Marett asserts that his Aboriginal Hobgoblin is not a bloodless wraith; but if we look simply at the vocabulary and imagery of these passages, is it too fanciful to detect the seeds of the idea of the Ringwraiths, especially their uncanny cries heard in the woods, their invisible (masked) faces, and their power to inspire terror? Consider, from ‘The Siege of Gondor’ in The Lord of the Rings, the following, and note the word shuddering, which could be taken as echoing its use by Marett:

Pippin knew the shuddering cry that he had heard: it was the same that he had heard long ago in the Marish of the Shire. (2004: v. iv. 809)

Another long screech rose and fell… Faint and seemingly remote through that shuddering cry he heard winding up from below the sound of a trumpet. (Ibid.)

These refer back to the earlier encounter (from ‘A Short Cut to Mushrooms’):

A long-drawn wail came down the wind, like the cry of some evil and lonely creature. It rose and fell, and ended on a high piercing note (2004: i. iv. 90). 

Note that this cry is a late element of ‘A Short Cut to Mushrooms’, put in (to replace the sound of hoof-beats) in the very last rewriting of the chapter during the so-called ‘Second Phase’, tentatively dated by Christopher Tolkien as probably before October 1938. Recall that Tolkien accepted the invitation to give the Lang Lecture between 8 and 14 October, and so might have been reading Marett’s essays at this time. Did the bullroarer, the Hobgoblin of the woods, help shape the idea of the  Ringwraiths?


Fry, Michele 2007 Oxford in Michael D. C. Drout J. R. R. Tolkien Encyclopedia Routledge.
Dundes, Alan 1976. A psychological study of the bullroarer Man 11 220–238.
Fison, Lorimer & A. W. Howitt 1880. Kamilaroi and Kurnai. Melbourne: George Robertson.
Haddon, A. C. 1898. The Study of Man. London: Bliss, Sands & Co.
Hammond, Wayne G. & Christina Scull. 2008. The Lord of the Rings: a Reader’s Companion. HarperCollins.
‘The Kettledrummer’. 1896. Judy’s Literary Kettledrum.  Judy, 19 Aug.,  717–718.
Lang, Andrew 1884. The bull-roarer: a study of the mysteries. In Custom and Myth. London: Longmans Green and Co.
Marett, R. R. 1910. Savage supreme beings and the bull-roarer. Hibbert Journal 8, 394–410. Reissued in The Threshold of Religion (1914).
Marett, R. R. 1914. The Threshold of Religion. Second Edition. London: Methuen & Co. Ltd.
Marett, R. R. 1929. The Raw Material of Religion. London: Oxford University Press.
McIlwaine, Catherine 2018. Tolkien, Maker of Middle-earth. Oxford: Bodleian Library.
Robertson Smith, William 1889. Lectures on the Religion of the Semites. Fundamental Institutions. First Series London: Adam & Charles Black.
Scull, Christina & Wayne G. Hammond 2006. The J. R. R. Tolkien Companion & Guide: Chronology HarperCollins.
Smith, Arden R. 2000. Possible Sources of Tolkien’s Bullroarer. Mythprint 37, No. 12, whole no. 225 (December).
Tolkien, J. R. R. c1967. Nomenclature of the Lord of the Rings in Hammond & Scull (2008), pp. 750–782.
Tylor, E. B. 1881. Review of Kamilaroi and Kurnai. Academy 19, 264–6.
Tylor, E. B. 1884. Special session of the Anthropological Society of Washington, 11 October. Transactions of the Anthropological Society of Washington 3.
Obituary of R. R. Marett. 1944. Oceania 14.