The Nandi bear

The full file

The Nandi bear entered print through the Journal of the East Africa and Uganda Natural History Society. Between 1910 and 1914 its contributors described a heavy, shaggy animal of the Uasin Gishu plateau and the Nandi country of western Kenya. They gave it the Nandi name chemosit, and some of them were sure it was new to science. No specimen was ever collected, and the identifications that followed all named animals already known.

What was reported

The founding account is Geoffrey Williams' note "An Unknown Animal on the Uasingishu", published in July 1912. He gives no date for the sighting, saying only that it was several years earlier, just after the Nandi expedition, the British campaign of 1905–06. That is where the conventional date of 1905 comes from. Walking ahead of his safari in mist near Sirgoit Rock, Williams and his cousin saw a large animal sitting up on its haunches only a short distance away. He thought it nearly five feet high, with thickly furred forequarters, almost bare hindquarters and a long, pointed head, and he wrote that its posture was like that of "a bear at the 'Zoo' asking for buns". He fired and missed. He later learnt that the Nandi called the animal chemosit, and a Nandi servant shown the Jungle Book pointed to Baloo.

In July 1913 C. W. Hobley, a senior colonial administrator, gathered further reports in "On Some Unidentified Beasts". Major Toulson, a settler, saw a black animal 18 to 20 inches at the shoulder with a shuffling, bear-like gait. In March 1913 N. E. F. Corbett, District Commissioner at Eldoret, almost walked into an animal by the Sirgoi River that was reddish-brown, larger than a hyena, with largish ears and a white streak down the hindquarters. In the same issue G. W. Hickes, engineer in charge of building the Magadi Railway far to the south-east, described an animal he passed on his motor trolley on 8 March 1913. At first he took it for a hyena: tawny, shaggy, high at the withers, short-eared, with no visible tail.

The evidence and where it is held

What survives is paper: the journal itself, now digitised from library copies on the Internet Archive. Its one exhibit is a woodcut of a footprint sketched by an engineer, F. Schindler, from mud beside the railway pipeline. Hobley's caption places it at Mile 74 of the Magadi line; Hickes puts the pipeline outflow at about Mile 71. A skin at Kabras was never obtained, a Boer's kill was never found, and the animals the Nandi and Marakwet were said to have killed left no remains. No skin, skull or bone is known in any collection.

What investigators concluded

The contributors disagreed among themselves from the start. Around 1910 Hobley thought the Nandi forest stories might be explained by a newly reported potto, a small primate whose head was said to look like a bear's. In 1913 he noted that sceptics proposed the aardvark. In August 1914 A. Blayney Percival, the game ranger, wrote that he was "still a bit doubtful" about a new animal. He thought the chemosit a composite of several known animals, among them the ratel or honey badger, the leopard and the baboon, kept alive by native legend and by settlers' habit of putting any unrecognised animal down to it. He also treated the Nandi and Magadi reports as different animals. The Lancet reviewed that issue in October 1914 and called the description "by no means a scientific or illuminating" one.

In 1927 the ethnographer G. W. B. Huntingford set the Nandi tradition out apart from the animal. To the Nandi, the chemosit was "a devil with one leg and nine buttocks and a fiery mouth" that caught children at night; the same name had been given to a fierce animal reported across Kenya, which he credited a Dr Heard with calling the Nandi bear. He knew of few, if any, authenticated European sightings. According to later reference works, R. I. Pocock identified the reports as spotted hyenas in 1930, the Natural History Museum in London said the same of many reports in 1932, and G. G. Simpson wrote in 1984 that most, if not all, were ratels.

The conventional explanation

The leading explanation is misidentification of known animals, chiefly the spotted hyena, with the ratel for the smaller reports. Its strength is that it rests on the witnesses' own words. Hickes first took his animal for a hyena; Corbett compared his with one; Percival had settled a similar Sotik beast with a picture of a ratel. The sightings were brief, at dusk, in mist or at speed. Its weakness is that one known animal cannot cover them all. Williams' animal, if anywhere near five feet sitting up, is far larger than a ratel, and Corbett's reddish coat and white streak suit no hyena well. Andrews' 1923 proposal of a surviving chalicothere, later taken up by Louis Leakey, is no longer argued: chalicotheres were herbivores.

What remains unexplained or disputed

What Williams and his cousin saw cannot now be settled. Nothing was ever collected that could be tested. The reports came from places hundreds of miles apart, from people who already knew the story. Whether the Nandi tradition grew out of a real animal, or settlers fitted animals they did not recognise to a Nandi name, is a question the record cannot answer.

Sources

• Journal of the East Africa and Uganda Natural History Society, vols 1–2 (1910–1912) — Internet Archive scan. Williams, "An Unknown Animal on the Uasingishu", vol. II no. 4, July 1912, pp. 123–125; Hobley's "Kavirondo Potto" note in vol. I no. 2.
• Journal of the East Africa and Uganda Natural History Society, vols 3–4 (1912–1914) — Internet Archive scan. Hobley, "On Some Unidentified Beasts", vol. III no. 6, July 1913, pp. 48–52, with the footprint woodcut; Hickes's Magadi note, pp. 53–54; Percival, "The Chemosit", vol. IV no. 8, August 1914, pp. 127–128.
• The Lancet, 31 October 1914, vol. 2 for 1914 — Internet Archive scan. The review of Percival's note under "Journals and Magazines".
• Journal of the Royal Anthropological Institute, vol. 57 (1927) — Internet Archive scan (doi:10.2307/2843712). Huntingford, "Miscellaneous Records Relating to the Nandi and Kony Tribes": the chemosit as night demon, and the animal reports kept apart from it.
• Nandi bear (archived) — Wikipedia, used as an index only. The citations for Pocock 1930, Game Animals of the Empire 1932, Simpson 1984, Andrews 1923 and Leakey 1935, which this file could not open directly.