In the great sanctuary of Wolfish (Lykios) Apollo at Argos a fire was kept burning which the Argives called the fire of Phoroneus. Sir James Frazer1
The Fire list is very different.
Having used a variety of wordlists to solve various puzzles connected with words I noticed that 'fire' words appeared in all of them and I was curious to see how they fitted together, if they did. The choice of a root seemed an easy matter. The entry for E. fire in the Oxford Etymological Dictionary gives as cognates Gr. pur, Um. pir, Cz. pyr, Arm. hur, and Toch. por, which points to BR or CR as the root. Without looking further than fire itself I could think of brand, braise, bright, broil, burn, fire, fry, and fury, as well as char and sear. Then I found that E. pile and E. pyre both mean 'a heap of combustibles for cremating a dead body'. So I allowed for other examples of L/R equivalence and included BL and CL. This produced what still appeared to be a manageable list of roots for a Fire list.
BR, FR, etc; BL, FL, etc; CR, GR, SR, etc; CL, GL, SL, etc.
On this basis I began to collect words, following the rule that eligible words must fit the roots but might have any meaning. Within a few hours I realised that this was an impossible task. These roots cover not hundreds but thousands of eligible words. It would be possible to collect all these thousands of words and organise them into categories but the effort would only confirm what had already become evident: that most words in most European languages derive from these roots. In other words, most words in most European languages derive from 'fire'. This would be remarkable if true. I wondered if this could be demonstrated in some more manageable way.
The method I adopted was to reverse the normal rules. Instead of collecting words that fitted a root and then allocating them to categories according to their sense, as in all previous lists, I collected only words which related to fire, and then tried to fit them to these roots.
The theme-driven approach should not work. It should not produce a pattern, let alone what appears to be a universal pattern. If we were to collect the names of animals or settlement features, we would get a random list of words with no discernible pattern. But fire is different. The pattern that emerged suggests that most words in most languages descend from a root meaning 'fire'.
The search involved reading through many dictionaries and eventually produced almost 1300 words which are in some way related to fire. They were then tested against a structure based on the 'fire' roots laid out above. I found that the primary monosyllables, B, P, F etc. are rare as independent words but very common when combined with one of the six TEC consonants, notably with L and R (5 and 6 below). A further phase of evolution is shown by longer words based on the same roots. Another remarkable finding was that every category in this theoretical scheme has its share of words.
STRUCTURE OF THE 'FIRE' LIST
0. Primary roots: monosyllabic words for 'fire': B, P,F, C, G, S,H,(-) etc .
1. One of the primary roots combined with B to give BB etc.
1.a Further compounds of (1).
2. One of the primary roots combined with M to give BM etc.
2.a Further compounds of (2).
3. One of the primary roots combined with C to give BC etc.
3.a Further compounds of (3).
4. One of the primary roots combined with D to give BD etc.
4.a Further compounds of (4).
5. One of the primary roots combined with L to give BL etc.
5.a-f Further compounds of (5) in six separate lists.
6. One of the primary roots combined with R to give BR etc.
6.a-f Further compounds of (6) in six separate lists.
To understand what these roots represent we have to go back to the TEC.
New sounds
The TEC gives us the capacity to reduce all the sounds in modern European languages to six working consonants. It represents a common level basic to all these languages. The latest possible date for this common level is the Upper Palaeolithic, before permanent settlement, when uniform culture implies a uniform language. It might be older but is unlikely to be more recent. Pan-European words such as cow, dance and fire belong to this postulated Palaeolithic language but many of them have wider (and older) links. The uniform culture of the Palaeolithic came to an end in the Mesolithic when we began to settle permanently in our present locations. As populations became isolated and culturally self-contained, the common Palaeolithic language evolved locally into the different but related languages of Europe. Palaeolithic sounds also evolved locally into a greater range of sounds, to the point where every national language is now defined by unique sounds which are difficult for foreigners to master.
Several factors may have contributed to this expansion. There is, as mentioned, the tendency for any isolated language to evolve in a unique way. This is true down to the level of a rural village. The dialect of Polmont, a suburb of Falkirk (West Lothian), was recognisably different from the dialect of Brightons, an adjoining suburb, and both, needless to say, were different from Falkirk. The increase in variety served the need for increased lexical resources as human culture becomes more complex. Thirdly, it seems that unusual sounds may serve the ends of national or tribal identity. A further factor is the contribution made by learned men towards inventing and improving alphabets.
These processes can be traced in the range of consonants in use today. The Finns, living a quiet life in the northern forests, achieve phonetic exactitude with only 9 consonants (D K L M N P R S T) and 4 aspirates (H J V Y). Irish Ogam and Runic had only 12 consonants. Irish now has only 13 consonants but compensates with aspirated sounds and complicated spelling. Runic lacks graphs for D, G, P, and V/W but made do with T, K, B and U. Greek began with 12 consonants but, after several learned revisions, ended up with 18. Cyrillic is based on Greek but needs 22 consonants including 9 aspirates. Arabic uses 27 consonants including 8 aspirates. All this pales beside the complexity found in the Caucasus where Abkhaz has 56 consonants or regular consonantal clusters, Bzyp, a dialect of Abkhaz, has 67, and Ubykh, recently extinct, needs 80 consonants for exact transcription. Since the way of life in the Caucasus is not markedly more complex than that of Finland, perhaps these very complicated sounds serve to preserve a local or national identity in an area proverbial for internecine warfare.
Disappearing Consonants
These complex European (and some Asian) sounds are the end-result of a long period of evolution whose roots lie in a common language spoken in the Palaeolithic. But language did not begin to evolve twelve thousand or even fifty thousand years ago. When we reverse the process of consonantal expansion in Europe, this great variety shrinks back into the six consonants of the TEC but the TEC is already an evolved and complex situation. It is not the start of anything, only a stage in a very long process.
Most of what we need to know to understand this earlier process of evolution and how it relates to the Fire roots has been discussed in the first chapter. The TEC itself tells us something, for B, C and to a lesser extent M are evidently old sounds, with flourishing tails of offspring which point to their age, while D, L and R have few if any relatives and must be more recent.
1 B MB P F V W
2 M Mb N Ng Gn F V
3 C G J K Q S X Y Z Ch Sh H
4 D T Th
5 L LL
6 R RR
This can be shown in another way. As we saw in the first chapter, several of these basic letters are related. M arose as Mb, an offshoot of B, while B and C were once a single sound. L and R are unstable sounds which were often confused and which still, according to the well-known champion of human rights, Lauren Oda, appear and disappear for no evident reason. The Armenian equivalents (discussed in Chapter 1) include H’ for L and Gh for L and R, as in Arm. ghampar for E. lamp and Arm. h’amr for E. lame. This suggests that L/R was once a soft guttural sound, as Welsh LL and the Parisian R still are. Though this is by no means the whole story, L and R can be left on the sidelines as a side-shoot of C. This reduces the TEC to two effective consonants: B/C and D/T.
Evidence from many parts of Europe shows that T and S or C regularly overlap. .
- C for St. In written Croatian C represents the sound St.
- St for Sc. Lat. stincilla, scintilla ‘spark’.
- T for C. In transcriptions of Gaelic place-names, C/T and Ch/Th are sometimes confused. Johnston attributes the interchange of C and T as in Avich, 1297 Awath, to 'scribal error' but the scribe may have had difficulty in deciding between Ch and Th. Early forms of Athole include Athoclach for *Athoclath. Elliot (Arbroath) was Elloch or Elloth. Arbirlot (Arbroath) was Abereloth, Aberelloch and Aberellot. Monikie was Monichti and Monichi.
- T for Ch. In English T has the value Ch in capture.
- T for K. Arm. kark for E. cart.
- T for S. In certain Greek and Latin roots, S changes to T/D: Gr. ous, otis 'ear'; Lat. pes, pedis 'foot'; lis, litis 'strife'.
- T for S. T has the value S in E. inertia.
- T for SP. The river Spey is Tvesis in Ptolemy c.150.
- T for ST. French words containing ST are regularly reduced to T: tempest > tempête; priest > prêtre; conquest > conquête.
- T for ST. E. tinsel for Fr. estincel (11c), E. tin for Lat. stannum.
- T for TS. In phrases such as G. an t'snaimh 'at the swimming place', T eclipses S. Kintail is G. cinn t'saile.2
- Th for CT and ChT. Watson notes that British or Latin act, oct, uct, ect and ict give aeth, oeth, wyth, eith and ith in Welsh; hence W. caeth, G. cacht 'slave'; W. noeth, Ir. nocht; W. rhaith, Ir, reacht, Lat. rectum.3
To identify T as a form of C or S completes the process of reduction. If D/T is an offshoot of C/S, the six consonants of the TEC eventually merge into a single complex sound combining features of B and C. This is the Heathcote sound that we met already. With Heathcote we have moved as far back in the history of language as we can. We have reduced language to a single syllable, an indeterminate blowing sound which has been transcribed as P, F, FF, Chi, Gh, Th and Wh. This is the first word, and it meant 'fire'.
Working in the opposite direction, the process of evolution began with the Heathcote sound. It is perhaps an oversimplification to claim that from this single sound all the other more defined sounds have evolved but the consonants we know today have certainly grown from this base and established their autonomy in regular use over the aeons of prehistory. Hs, which we can equate with fire, first gave us the important B cluster and the even more important C cluster. Then, at some point a variety of B branched off to become M (which is easier to pronounce) while C or Ch became T or Th. L/R may have started as a guttural Gh. Since sounds do not travel on their own it follows that all our words derive ultimately from the Heathcote word, which meant 'fire'. Fire came before everything else. The proof of this is given by the words themselves.
The Fire List.
The link with the Fire roots is shown by the fact that most of the Fire words begin with B, C or their variants. The earliest and most primitive words are the monosyllables Bi, Fi and Chi which are shown as primary Fire roots in the first row of the table. This first row corresponds to the varied sounds of what we might call the Heathcote cluster. Very few Fire words begin with D. None at all begin with R but L is found in 50 words with the root LM and 42 words with the root LC. L in these roots may represent Gh or Ch, as proposed in Chapter 4. As L and R appear to have run in parallel, it is unlikely that one could have a history independent of the other. The only plausible explanation for the absence of RM and RC words from the Fire list is that L was the original sound, and that it ran alone for a long time before producing R as a variant.
When discussing ways of avoiding error in a word-list, I advised against including words with initial vowels. This is no longer necessary in the ‘fire’ list since the criterion for inclusion is the meaning of the word, not its structure. Thus, G. breo ‘fire, flame’, Ru. par ‘steam, vapour’, Ru. var ‘tar, wax’, E. char, G. gor ‘light, heat’, E. fire, Fin. huuru ‘vapour, steam’, Nor. åre ‘open hearth’, E. ire ‘fury, burning rage’, Heb. or ‘light’, G. ur 'fire', and Lat. uro ‘to burn, destroy by fire’ can all be listed on equal terms under BR. E. apple falls into the same class as G. éibheall ‘flame, live coal’ and Gr. kapplos ‘dry stalks, twigs’. Gr. aithein ‘to light up’ and G. aodh ‘fire’ fall together with E. heat and E pitch. This allows us on occasion to identify the original form of a word: thus Gr. ipnos ‘oven, furnace; lantern’ can be compared with Gr. kapnos ‘to make smoke, to light a fire’.
On the other hand we should try to exclude prefixes. Smoke can be a positive or desirable aspect of fire when it is used for preserving food or sending a signal. But sometimes smoke and darkness are negative concepts as in Gr. dnophos ‘darkness’ and its cognates knephas ‘darkness, dusk, gloom’, nephos ‘mass of cloud’, and psephos ‘darkness, smoke, mist’. In every case the second element is Gr. phos ‘light’. Dno, Kne, Ne and Pse are negative prefixes which in these words define an absence of light. On the other hand, Ger. dämmerung ‘twilight’, Du. damp ‘vapour, steam, smoke’, Du. dempen ‘to smother a fire’, E. dim ‘obscure’, Ru. tyemnata ‘darkness’, Ru. dam ‘smoke’, Ru. damka ‘mist’, Lith. dumai ‘smoke’ and Alb. tym ‘smoke’ have been included as they appear to be neutral.
To supply words for all the new developments associated with fire, B, C and their variants have combined with every consonant in the later TEC. This matches the history of fire itself. It was known and controlled already by Homo Habilis but the great inventive expansion in its use took place in the Upper Palaeolithic, from 50,000 BP. An unexpected feature is that BL and BR are by far the commonest roots, supplying between them two-thirds of all Fire words (450 BL words and 365 BR words). This may be because L/R added a directive or locative meaning to the primary root. I have listed BL words separately from BR words but their numbers are quite evenly balanced, which tallies with the proposal that through most of their history they were a single sound. There are very few words with the structure BRL or BLR - such words could evolve only quite recently, after L and R became distinct sounds. There are virtually no words with the roots BLL and BRR, supporting the rule that duplicated consonants in the body of a word do not survive in speech - an example is Cansity for Kansas City.
Exceptional cases
We then get on to the interesting odd ones. A small number of Fire words begin with M, N, SM or SP. With few exceptions, the SM and SP words are found in languages in north-west Europe, including Britain, a distribution which brings us again to unusual sounds evolving in isolation. They are not the beginning of a trend: Chinese still has conventional Fire words like bizi ‘fire-grate’, huore ‘burning hot’ and chi ‘flaming, ablaze’.
Since it can be removed without causing damage, the initial S appears to be a prosthetic letter - a local speech pattern or lisp which has been copied into the written language and thus acquired validity. If the S is taken away this leaves us with initial M or P which both file under B. Initial SM, SN, SL and SR in Gaelic are found as M, N, L and R in other languages: thus G. sruth 'stream', W. rhwd, Fr. ruisseau; G. smear, W. mar, E. marrow; G. sneadh, E. nit; G. slais, E. lash. Further examples are shown below.
With and without prosthetic S
smelt (E.), Norw. smelte, Du. smelten ‘smelt, melt down’. > E. melt.
smeulen (Du.) ‘smoulder’. > G. maolan 'beacon'.
smoke (E.) ‘visible fumes given off by a fire’. > E. cook, bake.
smoulder (E.) ‘to burn slowly, giving off smoke’. > E. moulder 'to waste away gradually'. G. beoll 'fire'.
smùdan (G.) ‘smoke raised as a signal’; smudge (E.) ‘choking smoke; fuel for obtaining smoke’. > G. mùig 'darkness, gloom', E. muggy 'warm and damp, of weather', O.Nor. mugga 'a mist'.
spark (E.) ‘a glowing particle’. > Ru. margat ‘to blink, wink’, E. bark (used for torches).
speur (G.) ‘star’. > E. pyre 'pile of combustible material for burning a dead body'.
spingseti (Lith.) ‘to glimmer, twinkle’. > Gr. phengos ‘light, splendour, lustre’.
splang (G.) ‘sparkle, flash, blaze’. > G. flann 'red', E. flame.
splinter (E.) ‘a slender fragment of wood used as a torch’. > Crt. planuti ‘to catch fire’.
spodos (Gr.) ‘ashes, embers, wood ash; ashes of the dead’. > Gr. photos ‘light’.
spor (G.) ‘flint, tinder, flint or quartz strike-a-light’. > E. pyre (as above).
sprank (Du.) ‘spark’. SPRN > BRM. > Du. brand 'fire'.
sprokkel (Du.) ‘dry sticks’. SPRK > BRC. > Du. brok 'lump, piece'.
spruan (G.) ‘brushwood, firewood’. > E. branch.
svi (Norw.) ‘burn, singe, scorch’. > bi in G. bibidh 'large fire'.
svyet (Ru.) ‘light’. > E. heat.
Finally we come to a small group of words, eight per cent of the total, which begin with D, T or St. The majority are again from Eastern Europe and Britain. Since we can now equate T with St, and St with Sc, we can construct an equation which allows us to reconcile these local spellings with our standard categories and which incidentally provides a link between TEC 3 and TEC 4.
C, S, Z < > Sc < > St < > T, D
This allows us to recognise some strange local variants as standard TEC.
djegur (Alb.) ‘burn, scorch, scar, scald’ > CCR.
dzhjar (Ru.) ‘heat’ > CR.
dziuti (Lith.) ‘to dry, become dry’ > CT.
tvilkyti (Lith.) ‘to scald’ > CLC.
zidinys (Lith.) ‘hearth, home, fireside, focus, centre’ > CDM.
zjarr (Alb.) ‘fire’ > CR.
zvilgeti (Lith.) ‘to shine, be glossy’ > CLC.
Similar substitutions are found in the dialects of Ancient Greece, in which D was sometimes used in place of B, G, St, K, L or S: Gr. kaio ‘to burn, kindle, light up, set on fire, scorch’ is also found as daio. L is the only anomaly in this list. Its inclusion may reflect its early use to represent a guttural Gh or Ch sound, as suggested elsewhere. G. geal or giol 'zealous' (ie, burning like fire) is also found as deal.
The evolution of the TEC consonants can be summarised very simply.
Hs > B and C; B > M/N; C > T/D; C > Gh > L > L/R
Does the Fire list show internal logic? Very much so. Despite the range of sounds represented, similar words with identical structures from remote parts of Europe fall together in a very satisfactory way. The lexical ordering is convincing and suggests that the Fire list matches the reality of lexical evolution. All but two of the theoretical roots match actual words, generally in considerable numbers.
The Fire list does more. It expands our view of prehistoric society by emphasising the primary importance of fire in the evolution of human culture and, in consequence, of humanity itself. Among much else it shows us that hunters and herders used fire to improve grazing, to drive animals, to pen them in, and to protect them. It provides us with some very early metaphors in which ‘fire’ words are used to describe an arrow which was as sharp and as fast as a ray of light and a berry which was as round and red as a fire. It shows that women and fire are coupled in almost every northern society.
The age of these names is imponderable. Those who settled Europe fifty thousand years ago must have brought many older words with them, for they could not live outside Africa without fire. Our interaction with fire goes back in a direct line for more than two million years, and in the world of fire there are very few innovations. Fire, flame, sparks, smoke and ashes are all exactly the same now as they were two million years ago. The earliest built structure was the hearth, which was invented some fifty thousand years ago. Today our houses and even our caves, caravans and tents are equipped with chimneys, stoves, furnaces and ovens. These are new inventions but their names are not new words. W. llumon now means 'chimney' but originally meant ‘smoke’.
In the beginning was the Word, and the Word was Fire. Given that Gr. logos ‘word’ is a pun on Gr. phloks ‘flame, blaze’ and cognate with E. log, the Evangelist is likely to have been aware of this. An evangel is also 'fire'.
The interaction between fire, language and evolving culture coincides with the evolution of inventive humans with large competent brains who lived and hunted in large efficient social groups. The men could carry fire wherever they wanted, because their hands were free to do so, and their women could rear their infants in safety beside the family fire. Everything else followed. This brings us to the link between the woman and fire.
Words Linking Fire and Women
It was not only the Vestal Virgins of Rome and the virgin nuns of St Bride who tended a perpetual fire. Once upon a time all women did so, only they were not condemned to perpetual chastity. Many of these words have sexual implications. This is not yet ritual prostitution, but these words suggest that the maid beside her fire gave herself freely to hunters. This is to take the sense literally. But fire was also symbolised by the Scarlet Woman who also gave herself freely to hunters. She might also be periodically raped by the Black Man, a creature of brushwood and coal, who was also, from some different perspective, her son.
Edit:
The Cailleach represents the most archaic imagery of fire, which was, once, a female preserve. The most striking examples of women’s links with fire are of course the ritual hearths maintained by unmarried women. The best known are the Vestal Virgins who tended the perpetual fire at Rome but the nuns of St Bridget in Ireland also tended a perpetual fire which burned into the middle ages. There are many etymological links between words for ‘woman’ and words for ‘fire’. The first element of most of them is some form of fi, the elemental ‘fire’ word, which represents the blowing noise made to encourage a spark to catch on tinder.
Check: G. bàn ‘white (fire); woman’
G. brann ‘firebrand, woman’,
G. feim, Fr. femme ‘woman, female, wife’
G. guin ‘woman’, cognate with feim and with E. whin (used for kindling).
G. piuthar ‘sister’.
Lat. virgo ‘virgin’, an unmarried woman who stayed at home.
E. whore.
E. harlot.
E. female.
E. woman.
E. wife.
purdah
G. eibh ‘fire’, cf Eve.
Lat. virago
bibi (Ar.?) 'woman', literally 'fire-fire'.
brann (G.) ‘woman; firebrand, burning coal’.
cailin (G.) 'girl, damsel, maid'. G. cail 'burn'.
cailleach (G.) G. cail 'burn'; ceal 'vulva'. The Cailleach was the Goddess.
cé, céile (G.) ‘spouse’. *ci is a primitive ‘fire’ word.
daughter (E.). G. dòth ‘scorch, burn’.
feim (G.) ‘woman, female, wife’: ‘she who is concerned with the fire’. *fe is a primitive ‘fire’ word.
female (E.). *fe is a primitive ‘fire’ word
femme (Fr.) ‘woman’.
filia (Lat.), FL ‘fire’, like E. fuel.
firinn (G.) ‘maiden, girl’. FR ‘fire’.
girl (E.). CR ‘fire’.
guin (G.) ‘woman’. Cognate with G. feim.
hag (E.). From the same root as Skt. agni ‘fire’.
harlot (E.). HR 'fire'.
lady (E.) ‘a woman who tended a fire’.
maid (E.) ‘young unmarried girl’, as G. maide ‘stick’, Nor. ved ‘firewood’, E. wood.
piuthar (G.) ‘sister’ Pi is a basic 'fire' root.
sister (E.), a cognate of G. piuthar.
virgin (E.). FR ‘fire’.
whore (E.). FR ‘fire’.
zhena (Ru.) ‘wife’, related to Ru. zhenye ‘burning’.
Hot words
In setting up this list I made only one assumption, amply justified, which is that the element found as Pi, Fi or Chi means 'fire' or 'hot' and that its suffix, -L or -R, defines a place or an object. Thus, E. fire means 'hot place' or 'source of heat'. The elemental Pi or Fi may be the sound made when blowing on sparks or hot embers to generate a flame. It is no longer common in its original form but very common indeed in compounds.
It may seem improbable that E. pee and Fr. pipi mean ‘hot’ or ‘hot-hot’. No-one living today in Britain or in France would give it such a name. A liquid at body temperature is only 'hot' if one lives in a very cold climate. But then not so very long ago we did. Those parts of France and Britain that were not covered by ice were covered by tundra. In the summer it was green and lush, like an Alpine meadow, but otherwise France and Britain were not unlike Greenland today. And in Greenland today, where temperatures may reach minus 70 degrees, a warm liquid is at a premium. Urine is particularly valued for making instant repairs to the ice covering on sled runners since it freezes immediately to form a smooth, tough surface.4
Other names for bodily fluids have a similar origin. E. urine and G. fual ‘urine’ are also ‘hot’ words, the first being an eroded aspirate. Gr. peos ‘penis’ may mean ‘source of hot stuff’. E. blood is known in Gaelic as fuil.
There is a curious story about the use of blood as hot stuff in a very cold climate. Fionn and the Feinne were the legendary deer-hunters of the Gaels. When Fionn and his companions went to Lochlann, somewhere in the North, they suffered enchantment. Their knives froze to their hands, their bottoms froze to their seats, and their seats froze to the earth so that they were unable to move. To release them from this enchantment it was necessary to capture the three daughters of King Gil and pour their blood over the captives. This worked like a charm and there was enough to release them all except Conan, who had to be pulled off his seat at the expense of his skin. The three daughters of King Gil survived the operation.5 This story of extreme cold is presented as a tale of magical enchantment to an audience which had no experience of extreme cold. Did a Norwegian visiting Ireland or Scotland tell his audience that he had used the blood from a captive animal – a reindeer or a dog – to cure frostbite? I can find no references to the use of blood for this purpose in the Arctic but it is not beyond belief that is was done. The use of fresh blood for magical (or religious) purposes is a standard theme. In the legend of Amys and Amelis, Amelis sacrifices his two young sons to cure his friend of leprosy by bathing him in their blood. The disease immediately vanishes, Amelis stands straight again, and the boys are found miraculously whole. Is leprosy in this case perhaps frost-bite? Whatever the truth of these legends, both fual 'blood' and fuil 'urine' are names for a warm liquid, as are E. blood, Fin veri 'blood', and Ru. krobh ‘blood’.
This is by no means all. As we might fear, E. crap, poo and shit are also ‘hot’ words. E. shit is paralleled by Heb. she’ten ‘urine’ and Du. schittern ‘to shine, glitter, sparkle’.6 Ger. Harn ‘urine’, Sc. sharn ‘cow-dung’, Hind. garm ‘hot’ and E. warm are all CRM words. E. milki, Ru. myelkat ‘to flash, gleam, glisten’ and Hung. meleg ‘warm, hot’ share the MLC root. W. llaeth, Fr. lait, and Lat. lac ‘milk’ are also ‘hot stuff’. G. blioch 'milk' is a form of G. blath 'warm'.
The pipi argument is equally true of all of these words. Only people living in a very cold climate would define substances at body heat as ‘hot’. The latest possible time for these naming events was the last glaciation in Europe, which peaked some 18,000 years ago but they might be much older and did not necessarily take place in Europe. In that case there might be unrelated names for blood and milk which were in use before the most recent episode of very cold weather. One such name is Lat. sanguis 'blood', a CM word derived from hunting. Gaelic has many different words for milk: bainne ‘white stuff'; cèo; finn ‘white stuff’; fìor; gall; lac; suth or segh (E. suck); tomhladh 'cow's milk'.
Some plant names also take us back to a very cold climate. E. plant itself has the same form as Crt. planuti 'to light a fire'. G. lus ‘plant’ is a cognate of Lat. lux ‘light’. To explain these links we must again go back to the tundra of the most recent glaciation. At no other time or place in Europe have plants served uniquely to produce heat and light. In the Arctic wicks for oil lamps are still made of moss, the pith of reeds and twisted lengths of hair (pith and hair are ‘fire’ words). When the climate warmed and trees spread over the landscape they also were seen primarily in terms of their combustibility and the names given to them reflect this: E leaf, E. plant, G. craobh ‘tree’ and Fin. lehti ‘leaf’ are all Fire words. Du. zwam ‘fungus; tinder, touchwood’ retains a direct link between a plant and lighting a fire. Fire remained a vital naming force in European languages for a long time. Post-glacial fruits such as E. apple, berry, cherry and plum were so-named because they resemble small red fires while E. bloom, flower, and lily are figurative names for objects which resemble flames.
Our proposed ‘fire’ word was also used to name a different kind of fiery substance. Pepper is widely known by two very similar names, both duplicated for emphasis. Bantu pili-pili and piri-piri, and Ar. filfil and falaafil are ‘hot-stuff-hot-stuff’. Arm. bi’ghbegh has the same structure and shows the equivalence of Armenian Gh and L/R. E. pepper, Gr. peperi, Lat. piper, O.Slav, pipiru, Lith. pipiras, and Fr. poivre all come from a variant meaning ‘hot-hot-stuff’. Nahuatl chilli is simply ‘hot stuff’. We find comparable (and very primitive) structures in G. bibhidh ‘very large fire’, Alb. xixe ‘spark’, Alb. zeze ‘black’, and Fr. pipi. W. pefr ‘radiant, beautiful’ is a worshipful epithet for fire, not for water. As a PNE, there are three small Pefferburns in Lothian (Scotland) and four more elsewhere. They all refer to former beacon sites such as that which once stood at Aberlady, an old port at the mouth of one of the Lothian Peffers. E. beck, brook, burn, and stream are all ‘fire’ words.
Old Oaths
Finally, did our ancestors swear by the Fire? The following exclamations are all related to E. puff 'breath' which is very near to the First Word.
Ancient Greece: Babai! Ea! Eya! Phû!
Ancient Rome: Eheu! Fu! Fufae! Papae! Phy! Vae! Vah!
Basque: Pufa!
Croatia: Fui!
England, on smelling blood: Fee Fi Fo Fum!
England: Ah! Fie! Phew! Pish! Poof! Pooh! Woe! Wow!
Flanders: Bah! Boe! Foei!
France: Aïe! Fi!
Germany: Pfui!
Iceland: Fúi!
India: Chi Chi!
Italy: Oibo! Ohi! Ahi!
Lithuania: Vai!
Norway: Fy! Fyrop!
Russia: Ev! Oova!
Scotland: Fich! Foof! Fuich! Hech! Peuch! Wow!
Sweden: Fy! Oj!
US: Whew!
Wales: Ffei! Hach!
Primeval syllables?
Having got down to the level of a meaningful primordial monosyllable I looked at some more early words. All the Heathcote consonants became early words. The disyllabic roots such as CR and DM consist of two elements combined. In this case the first is Co ‘together’ which is still in use as a prefix. Co belongs to a very early phase. As a new sound it must have identified a new concept and its appearance with the meaning 'to gather' must identify the time when our ancestors first began to use fire to trap and kill animals. CM and CR both mean 'gathering many' and mark the development of new skills which allowed our ancestors to round up a larger number of animals. With the appearance of such compound words human language entered a new phase.
The second element in these roots is represented by M, R or D. Can we identify these words? This might seem a very long shot but these roots are very old and very basic and they should be both durable and widespread. If Co survives, Mo, Ro or Do or some equivalents might also. We need look no further than Dwelly’s Gaelic dictionary to find five appropriate words. More surprisingly, Chamber’s English dictionary confirms three of them and adds one more.
Bo: G. bè 'life; wife, woman, female; night (fire)'. G. bo 'cow, fawn'; perhaps any prey.
Mo: G. mò ‘larger, largest; greater, greatest’, E. (obs.) mo ‘more’. CoM means ‘to gather a very great number’. The most important compound with Mo is MoRo, found very widely as *mar 'to hunt'. G. neo, E. no and Gr. mi are negative, impeding or diminishing words.
Co: 'together'. A very early element which is still in use as a prefix in English, in Latinised words and in Russian with the same meaning.
Do, To: G. do, to: ‘in the direction of, all the way in the direction of, in contact with, close against’, E. to: ‘in one direction, forward, into position, contact, closed or harnessed condition’, as in ‘put the door to’. E. do ‘to put, place’. G. do is a negative particle implying an obstacle or difficulty. CD means ‘to gather in a closed place’. It is more specific than the CM or CR roots and this may explain why it is less common.
Lo, Ro: G. ro ‘very much, exceedingly’, G. la or le ‘with, together with, in possession of’, in G. ro-mhath ‘very good’, G. ro-choill ‘a thick wood’, G. ro-mhar ‘a full sea’. Ro is cognate with Lat. re- ‘again’ and is the second element in G. mór ‘large’, E. more and E. very (for which Lat. veritas ‘truth’ does not offer a very satisfactory origin). CR also means ‘to gather a very great number’. The RB glossary suggests that the original root was not RB but ChRB. CR brings us back to the compound CoRo. It is the second element in Gr. makros ‘great’ and, with a negative prefix, Gr. mikros ‘little’.
This is compatible with the proposal, made on anatomical grounds, that the earliest words were simple syllables, such as Ba, Co, and Fi. We can even see how these primordial syllables were combined to supply the new words needed to define new elements of our material culture and abstract thought.
