Inter-specific Copying by Blackbirds
By Joan Hall-Craggs
This article was originally published in the Journal of the Wildlife Sound Recording Society Vol4 No7 Spring 1984
The title page of Richard Meares' "Bird Fancyer's Delight" (1717) includes the blackbird (Turdus rnerula) amongst species suitable for ".... ye Teaching of all Sorts of Singing-birds, after ye Flagelet and Flute ....",so it was already recognised as a vocal mimic of an alien species, albeit no other than Homo sapiens in combination with some of his upper wood-wind.
Yet of the 43 melodies presented in musical notation as appropriate models for 11 imitative species (at least, when kept in captivity) none is recommended for the blackbird. One would like to believe that its expertise was considered to be above such mundane tuition for, without doubt, it has a formidable capacity for learning and remembering through the years the sound patterns of both its own and other species.
J.M. Bechstein, a doctor and an authority on the maintenance and habits of cage-birds in the late 18th and early 19th centuries, was probably the first to publish detailed accounts of the normal songs and imitative abilities of the many species which were kept in captivity by the German bird fanciers. In his "Cage and Chamber Birds" ( 1794, 4th ed. 1840) he wrote of the blackbird: "The song of the male is melodious and consists of deep sonorous passages .... though intermixed with others which are rather harsh ...... Its memory is so good that it learns not only to repeat words but to whistle several airs without confounding them together. His "Handbook of Cage Birds" (undated) includes the remark: "They whistle or pipe songs and marches in very superior style and may even be taught to speak". But he is less enthusiastic about the blackbird in the "Natural History of Cage Birds" ('1812, 3rd ed.), main-taining that the song is ".... not destitute of melody; but it is broken by noisy tones and is agreeable only in the open country". Maybe he had found that a blackbird's voice in a confined space with reflecting sur-faces could be overpowering; or perhaps, like myself, he had started to give more attention to the "rather harsh" soundswhich are quite often appended to the whistled sections of the song-phrases. Yet it seems not to have occurred to him that this vocally acquisitive bird, when kept in captivity, would be likely to reproduce in so far as it could, some of the noisy and diverse sounds commonly heard in a household, especially when largely deprived of the opportunity to model its repertoire on the sounds available in its normal habitat.
Witchell (1896) comments astutely on the story of a blackbird imitating the crowing of a domestic cock which ".... has often been repeated and prob-ably owes its notoriety to the want of observation in writers who mentioned it; for this bird readily reproduces other cries, although they are usually performed in a faint voice". He then attributes the blackbird with copying the voices of "Golden Plover, Greenfinch, Blackcap, Wood Warbler, Nuthatch,
Peewit, Swallow, Great Tit, Green Woodpecker, Goldfinch and Magpie". My observations support this list; four of the species are among the following illustrations; others are apparent to the ear and to the eye when analysed, but are not shown because of my failure to record the precise model for, not unexpectedly, blackbirds copy individuals and some of these may have died or moved elsewhere. The "faint voice" copies usually occur when the model sounds fall below, or far above, the blackbird's normal whistled range of c 1.5 -4.0 kHz; but other appropriations may be loud and barely distinguishable from the model, e.g. that of the great tit (Parus major) in Fig. 12.
In the first half of the 20th century no major studies on vocal mimicry were specifically concerned with the blackbird, but in 1960 Thielcke-Poltz and Thielcke noted that blackbirds regularly incorporate the calls of other species as well as mechanical sounds in their repertoires; for example, their kaspar hauser birds copied the sounds of tape recorder switches and the stridulation of crickets. One of their males learned the excitement call of a Peking robin and used it in the correct situation. The copying of human whistles is fairly common and Tretzel (1967) gives an excellent example of such copying combined with pitch transposition; he also noted the gradual degradation of the original signal with increased distance from the model whistle, as the phrase was passed from one individual to the next.
Anecdotes about blackbirds reproducing with fidelity all manner of well-known musical themes are too numerous to list here, but it is noticeable that nearly all of these relate to the classical, romantic and lyrical schools and 1 have yet to hear of any that come from contempor-ary, neo-classical or pop scenes. This may reveal more about the nature of the listeners than the selectivity of the birds! (One of my Oxfordshire birds sang jazz with tremendous verve despite the apparent impossibility of so-doing in the absence of a metronomically rigid beat which could be displaced and allow for stress on weak sub-divisions of a bar). The most popular model in the classical repertoire is the first subject of the last movement of Beethoven's Violin Concerto which has been reported from as far afield as Australia. I have never heard this sung by a blackbird but have no doubt of the accuracy of Len Howard's observations (1952) -she was her-self :+ violinist -and the reports of many correspondents. However, Miss Howard's bird seemed to work out the theme for itself by a gradual process of experimentation and modification, well illustrated in her book; and who is to say that Beethoven did not get the idea from just such a bird during his long countryside rambles? Maurice Ravel deliberately copied a black-bird far his leading brief melody in "Oiseaux Tristes".
But whatever the origin of musical themes, it is becoming increasingly obvious that bird song has provided, and continues to provide material for composers (Hall-Craggs and Jellis, in press) and there is no doubt that our ousel cock both gives and takes. Even when possessed of a large and varied repertoire it retains the ability and motivation to copy numerous sounds whether of musical or noisy content, and I think it is important to distinguish these two categories. Musical themes, whether learned from conspecifics or appropriated, are almost always used as introductory figures to song-phrases, while noises are reserved for codas (or terminal decorations as I tediously referred to them in 1962), Fig. 1. A rather rare example of an entire song-phrase derived from two different musical calls and one noisy one, all copied from a single alien species -the curlew (Numenius arquata) is shown in Fig. 2. The long 'cooorli' is placed first, then the equally tonal but brief 'cui cui cui' calls follow and the coda is a copy of the mild alarm call which may sound like 'pi-pi-pm' The copying blackbird shared a foraging ground with curlew but other blackbirds, lacking regular access to this ground, have copied probably from the original copier - the medial section (b) which is now common in the songs of the adjoining population and is readily learned by first-year birds.
Of course this orderly arrangement of material sometimes breaks down (although I have never heard a blackbird start a phrase with a true noise ranging over a broad band of unrelated frequencies) especially in the reper-toires of mature birds who, when creating long-duration phrases by combining short-duration originals, quite often retain noisy codas in their customary positions where they serve as connective material or even medial figures if they are sufficiently substantial. Such diverse copies as those of the magpie (Pica pica) and the short-toed treecreeper (Certhia brachydactyla) (Figs. 4, 3 and 21 ) have been treated in this way.
It was not until part-way through a long-term study of a small population of blackbirds around the L'Ecluse area of the Island of Sark that I realised that most of these codas -maybe all, if I knew more of the model calls and songs -were copies of other species; often faithful copies but sometimes modified or even, seemingly, parodied as in the final figure (f) of Fig. 2 which shows the curlew-based phrase wherein (f) sounds like a derisory form of (c). But such apparent parodies have to be heard to be appreciated. It is possible that the Sark birds have a special propensity for mimicry, but it is equally likely that I can now recognise the voices of species that were strange to me when I left London for the Oxfordshire Chilterns, moreover I am listening for alien voices in blackbird songs. There may be truth in both suggestions for in this area of Sark, differing habitat types such as agricultural land, woodland, gardens, cotils -l'.... the belt of predominantly dry heath encircling the island, extending inland from the cliff tops ... to join the central plateau" (Hountree 1974), stream and pond cluster together in as little as a quarter mile square; thus a bird may have a territory overlapping or bordering several of these and, consequently, have the opportunity to hear a wider variety of species than is usual.
Of the 31 resident species in Sark I have recordings of 20 copied by blackbirds; in some instances, particularly the great tit (Parus major) and the blue tit (P. caeruleus), copies by a single blackbird of several different calls and songs of each species. Amongst the remaining 11 are the shag (Phalacrocorax aristotelis), pheasant (Phasianus colchicus), raven (Corvus corax), crow (C. corone) and others which, supposing they had some appeal for T. merula, have voices which might tax even the beautiful and complex anatomy and physiology of the turdine syrinx. In addition Sark has interesting lists of winter visitors, passage migrants and an impressive list of accidentals, all potential models for the acquisition of strange calls as well as the songs of some of the summer visitors. For complete lists and all relevant information see Hountree (1974) "Birds of Sark", an essential work for any ornithologist who visits the island.
Most of the appropriations shown in the illustrations are taken from the repertoire of one bird, South-west. This is not due to the failure of others to produce equally good imitations but because S-W was recorded in the same territory through 5 successive years and, for this reason, his repertoire has been more thoroughly analysed than those of his neighbours, also l had better opportunities for recording his models because his territory included the house and garden of L'Ecluse where cassette recorders could be left running when I was elsewhere. Once acquired, some of his copies were retained throughout the years, others were kept but in modified form (Fig. 5) and a few discarded or forgotten. The true blackbird whistled song repertoire remained virtually intact but for reorganisation of the figures comprising the longer phrases and the addition of new material, some of the latter learned from new neighbours -even first-years -who moved into vacated adjacent territories.
The modification of material taken from alien sources poses the main problem in tracing models and one can only hazard a guess at how much may remain unrecognised if the original copies are not recorded. The models for many probable appropriations in my recordings have yet to be identified so a few examples of puzzling material have been left on the sound spectrograms with the hope that readers may suggest likely origins. Calls should be relatively easy to identify, songs are more troublesome because in most instances only a part of a song is copied. Most difficult of all are examples where several fragments of different species' songs are strung together with an occasional call interpolated. Also, as Tretzel found, the blackbirds will transpose (Fig. 6) although their copying range is great and sometimes exceeds 8 kHz. It is even remotely possible that the copiers themselves become confused when species such as the goldfinch (C. carduelis) and the linnet (C. cannabina) sing simultaneously and in close proximity.
Conversely quite obvious copies of the sounds of other species may be heard, even recorded, and no matching model recording made or found. Such is my recording of a blackbird two territories to the north of S-W who added the announcement call of a tom-cat to a song-phrase! Cats were seen hunting in this territory but none obliged by calling in my presence. But instantaneous confirmation of the source was obtained by re-recording the call 4 times, with appropriate intervals between calls, and playing it in a house just within hearing of an intact Abyssinian male cat. At the first call the cat leapt from its bed and stalked the sound source, stopped when the tape stopped, then on re-play made straight for the recorder. And how many of us have heard blackbirds and starlings (Sturnus vulgaris) as well as song thrushes (Turdus philomelos) (See Slater 1983) copy the trimphone bell when no recorder was at hand? Inspiration for one Cambridge blackbird was found in the "cross now" signal at a pedestrian crossing. I do not doubt that readers can supply many other such examples.
It would seem then that the blackbird is at least a potential rival to the starling and other accepted mimics. That it has not already been generally appreciated as such is probably due to the fact that, in contrast to the starling, for example, its specific song is so varied, powerful and musical, and the individual repertoires so large and diverse, that the "harsh noises", "chuckles" and other sounds added to the whistled sections of the phrases have often been ignored or dismissed as unworthy of note. This is understandable for, unless the individual's repertoire is familiar to the listener, attention is directed to the whistled song while the often quiet and sometimes brief inter-specific copies are dissipated in the general song or pass too quickly for recognition of the model species. Moreover, there are times, not always consistent, when the birds omit these codas; these may be (1) when the bird first awakens and sings from the roost, (2) during slow and sporadic delivery, often inter-spersed with preening, (3) when a rival male is approaching the singer or its territory and (4) in the late evening beating of the territorial bounds. This list is not exhaustive. Conversely, a blackbird's most prolific expression of mimicry may be heard during what I think of as 'rambling song' which, in its continuity resembles sub-song but is uttered in a loud voice. My few recordings of this have been made on quiet, warm afternoons, generally while the female is incubating. It is usually immediately preceded and followed by normal song.
Yet even when a repertoire is known and mimicry is anticipated it may be difficult to identify models especially when song, as distinct from calls, is copied; for entire songs are rather rarely copied, and sections heard out of context can be unexpectedly confusing. If a suspected copy is recorded but unidentified, it is helpful to re-record it in isolation or, if a spectrum analyser is available, to record the extracted section on that and let it replay repeatedly. Sonagrams are a great help but occasionally seem to mislead, especially when small, and to us, inaudible differences in timing are involved. (At normal playback speed on a Kay 6061B, c 12.3 cm = 1 sec.) And some latitude in fine detail is permissible; a revealing example of this is illustrated and described in Cherry (1957). In this respect too the blackbirds can deceive for some have a tendency to "tidy up"; see e.g. the magpie call (Fig.4), and the house sparrow (Passer domesticus) 'rattle' and song units (Figs. 8 and 9). Above all, in the field one is rarely, if ever, fortunate enough to record both model and copy at the same sound pressure level and with other variables constant, so perfect matching of visual displays is improbable.
The essential problem remains: why should a bird already so richly endowed with song commit to memory and reproduce such a wide variety of alien sounds? Perhaps the initial necessity to learn a sizable, though at the time fragmented, song repertoire from conspecifics combined with neurophysiological attributes which enable it to memorise and continue learning throughout life (or, amongst my recorded birds, at least until and including the 6th year) make it a compulsive vocal mimic.
Author's note: The functional definitions of types of avian vocal mimicry advocated in the careful review by Dobkin (1979) are not followed in this article because no attempt is made here to attribute function to any of the examples described or shown in the illustrations. Thus the terms copying, mimicry, appropriation and imitation all refer to inter-specific vocal copying. The apposite term 'appropriation' is taken from Dobkin and most of the examples shown here would fall within that category as he has defined it.
I am particularly grateful to Dr. Gerhard Thielcke who kindly provided me with original sound spectrograms of the rivalry calls of short-toed treecreepers. 1 am also indebted to F.J. Sellar and to L. Koch/BBC and V.C. Lewis/BBC for permission to use their recordings of curlew and bullfinch.
BECHSTEIN, J.N. 1794, 4th ed. 1840. Cage and Chamber birds. H.G.Bohn, London. Undated. Handbook of Cage Birds. ward and Lock, London. 1812. Natural History of Cage Birds. Groombridge, London.
CHERRY, C. 1957. On Human Communication. Technology Press of M.I.T. and John Wiley. New York.
DOBKIN, D.S. 1979. Functional and evolutionary relationships of vocal copying phenomena in birds. Z. Tierpsychol. 50: 348-363.
HALL-CRAGGS, J. 1962. The development of song in the Blackbird (Turdus merula Ibis 104: 277-300.
HALL-CRAGGS, J. & JELLIS, R.E. (in press). Birds in Music in The New Dictionary of Birds. ed. Bruce Campbell. Revised edition.
Fig 1 Typical organization of a Blackbird song-phrase. (a) and (b) whistled units (notes); both of these sections were sung in other contexts. (c) Coda comprising probable copy, or copies, of other species.
Fig 2 (a), (b), (c) are 3 different Curlew calls displayed on one sonagram. Below: multiple copying of a single species by Blackbird ECshowing how the 3 calls were combined to form a single phrase to which was added (f), an apparent parody of (c). (d) copy of (b) by SW in his 6th year, (e) a Curlew variant of (b).
Fig 3 (a), (b) Rivalry calls of Short-toed Treecreepers (sonagrams by G, Thielcke). Copy by Blackbird SW, also shown in Fig. 21. Note: Certhia brachydactyla, not C. familiaris, occurs in Sark.
Fig 4 Magpie call 'chai-ak' and copy by SW.
Fig 5 (a) Copy by SW in 1979 of final section of a Chaffinch song preceded by 2 units of a Kestrel call (c). (d) (b) modifica-tion of (c) (a) by SW in 1980. The structure and pitch of (c) were changed but timing and context retained. The terminal flourish in (a) was sung before the repeated units and the latter more closely resembled Oystercatcher piping which was audible from the cliff top,
Fig 6 Song Thrush units with cut-off where 8 kHz was exceeded, Downward transposition and compression of long unit by SW.
Fig 7 Same Song Thrush as in Fig, 6 singing overlapping but harmonic-ally unrelated frequencies. SW copy of this fairly common phenomenon attributed to independent control of the two internal tympaniform membranes in the syrinx.
Fig 8 House Sparrow 'rattle-churr'. (a) copy by SW. (b) SW copy of same species' calls; no good model recorded.
Fig 9 Centre: section of (continuous) House Sparrow song. Above and below: copies by SW.
Fig 10 Two segments of a Wren song, Copy by SW. The 'rattle' segment is a common model for Blackbirds and the copies are often louder and clearer than the reproduction shown here.
Fig 11 Bullfinch calls, Copy by SW.
Fig 12 Great Tit song in TP territory 1979 (a) Copy by SW in 1980 when the Great Tit had extended its territory to include part of that of SW (b) Copy by TP Blackbird in 1979 with the portamento unit ascending instead of descending (c).
Fig 13 (a) Great Tit song (beginning of bout). (b) copy by SW preceded by copy of single Blue Tit unit (c). (d) presumed copy, model U identified.
Fig 14 Beginning of Blue Tit song and copy by SW.
Fig 15 Blue Tit call and copy by SW.
Fig 16 Starling call when taking food to nestlings. Copy by SW followed by unidentified but probably copied units. These resemble Blackbird fledgling calls but no good models amongst sonagrams of these.
Fig 17 Starling whistle. song for which no Copy by SW followed by quiet, model found. unidentified
Fig 18 Greenfinch 'ripple' and 'zweeez' . Copy by Blackbird SH.
Fig 19 Greenfinch 'zweeez' in SW territory 1978. Copy by SW in 1978 (centre) Final version by SW in 1979, 80 and 61. (This recording 1980)
Fig 20 Goldfinch 'eez eez ...' Copy by SW while nest-building.
Fig 21 Goldfinch song (b) cf SW copy (b) below also terminal descending glissandi. (a) SW copy of Short-toed Treecreeper in different context from Fig. 3.