169

Chapter XXIII

Philology Remote Origins

Most cognomens, sur- or family names mean something. The majority of them derive or descend from common names or nouns. The name COMBS (see various spellings later) belongs to this group. In this study, therefore, I propose to consider the origin and development of the name COMBS, first as a common name, then as a place name, before I embark upon the arduous task of genealogical considerations. Such a Gargantuan task requires much dallying in the domain of philology and semantics. I confess at the outset that my labors in this respect have not been all-comprehensive, due, first, to the fact that the name COMBS is one of the mightiest in the study of philology and genealogy; second, that, as a common name, its remote origins lie hidden in a number of the Aryan or Indo-European languages, in Asia and Europe. Often and on, during the past thirty years, I have consorted with comparative philology, and with its blood-cousin, semantics in an effort to arrive at the source of the matter. The ardous [arduous] nature of the job, then, will become apparent at once.

From the viewpoint of comparison and neology, the task is slippery and evasive, full of pitfalls and false leads. One may easily find himself stalking an ignis fatuus, or heeding the Circean call of a deceptive Echo. Conscious of these temptations, I have tried to separate true etymology from "folk-etymology," to avoid jumping at conclusions merely because such-and-such "resembles" or "looks like" such-and-such. Likewise, I have avoided hypothetical conclusions, as far as possible.

Before building up my constructive, linguistic thesis, it is neces-

170

sary to resort to a little destructive logic, touching on the origin of COMBS. I regret to say that the methods of some of my correspondents over the United States have not been scientific. Among such theories I note: the Persian Kum, a sun goddess. Now, Persian is one of the Aryan languages, and Kum may bear some remote relationship to kumb'a, a Sanscript [Sanskrit] word which will be discussed later. The assumption is that Kum was a valley goddess, that she was born in a valley, or that she lived in a valley. (We shall see, later, that kumb'a, under the process of semantics, or change, becomes valley.) The trouble with this theory is that, in the days of the good Kum, kumb'a did not mean valley; at least, the sacred and literary writings of the ancient Hindus do not record such a meaning for it. It is true that Kum is also a town in Persia (Iran), and that it is situated in a valley; all lovers of a once famous brand cigarettes may be interested to know that the tomb of Fatima (favorite daughter of Mohammed) is in Kum.

Kuma is another word that has been suggested. It can be dismissed briefly as being a river in Russian Caucasia. The resemblance seems to be only co-incidental. The name is probably more or less ancient, and, like Kum, has no b. The few Slavonic words that I have found related to kumb'a have a b in them.

The Chinese Komei, and Japanese Kumari: Chinese and Japanese have little or no relationship with the Indo-European languages, and these words may be discarded without argument. Likewise, the East Indian Komabis. Such other names as the Sicilian comani, the Roman (Latin) Cumae, and the old Irish, or Celtic Coamb and Kyme also break down under the microscope of philology, since they possess no linguistic relationship with the various forms of kumba's, comb(e), etcetera [sic].

One of my correspondents made the startling discovery of an old Roman coin in England, on one side of which are the letters COM. She concluded from this that "King COM(bs?)" was one of the early British kings, during the Roman occupation. Now, the early Britons did not use coins as a medium of exchange, and Britian [Britain] had no kings under the Romans. "COM" is simply an abbreviation of the name of Commodus, one of the Roman emperors of the second century. It is not necessary to mention various other fanciful, linguistic flights.

I now proceed to my constructive thesis, one which I hope to be able to establish conclusively. Any attempt to establish the exact source and antiquity of the word from which COMBS derives would be a rash one, indeed. I have said above that "the remote origins lie

171

hidden in a number of the Aryan or Indo-European languages." I mean, here, of course, the first use of the word. Of one thing we may be certain: that the prototype of COMBS belongs to the Indo-European languages; that is, as far as we can determine from printed records, and from the various dialects. Before proceeding with our study, we may as well rule out of court the word comb (for the hair), and the French combe, a ridge, or hill - since these words descend from Sanscrit gambha, a tooth. The French word derives from Celtic comb, which also meant dyke and wave.

Since our study takes us back thirty-five hundred to forty-five hundred years, I mention briefly the language groups and languages which I have had to consider. All these belong to what are commonly termed the Aryan, or Indo-European languages.

SANSCRIT. The name is sometimes applied to the whole sacred and ancient language of India, to the Vedic writings, etc. Such writings date back to around fifteen hundred years B.C. Its locale was largely in northern and western India. It is probably the oldest of all the Aryan languages, since the race that spoke it extended farther eastward into Asia than any of the others. Records of its sacred literature are also older than any which we possess among the other Aryan languages. As we shall see later, our linguistic point of departure begins with the language.

CELTIC. This is also a dead language, dating back around twenty-seven hundred years. Its boundaries cannot be definitely fixed, but roughly, it was spoken in various dialects from the western European littoral, including the British Isles, through central Europe, and probably into Asia. Its present-day descendants comprise Breton, Welsh, Scottish, Irish, Gaelic, Manx, Cornish.

GREEK and LATIN.

The TEUTONIC or GERMANIC group: English, German, Dutch, Danish, Norwegian, Swedish.

The ROMANCE group: French, Spanish, Portuguese, Italian, Roumanian [Rumanian].

The SLAVONIC group: Russian, Polish, Czech (Bohemian), JugoSlavic [YugoSlavic] (Serb), Bulgarian.

In Asia, aside from Sanscrit: Persian (Iranian), Armenian, and probably some dialects in northern and western India. I make no

172

mention of hundreds of dialects belonging to the various languages mentioned above.

SANSCRIT. On the title page of this book there is a strange jumble of characters. Disentangled and transcribed into modern English spelling, they spell: kumb'a (or kumbha). This is the word the ancient Hindus used (fifteen hundred to twenty-five hundred years B.C.) for urn, pot, vase, trough, and grain measure. They seem to have applied it to anything hollowed out. "Anything hollowed out" - that is the phrase which we shall have to keep in mind during the rest of the discussion; it is the crux of our thesis. For, as far as the records go, kumb'a is more ancient than any other similar or related word of similar meaning in any other of the Aryan languages. Franz Bopp, in his Comparative Glossary of Sanscrit, gives the following definitions of the word: (a) vas aquarium, urna; (b) modii frumentarii genus; (c) tumor in superiore parse frontis elephanti. The third definition, a bump, a tumor, has survived only infrequently, as we shall see later. Boisacq, in his Dictionnaire etymologique de la langue grecque, lists the Sanscrit word kumb'a (acute accent over the u) [kúmb'a], with the meaning of a sorte de coiffure de femme; it is probably the same word as kumb'a, and, by extension, a "lump," or "bump" of hair. Kumb'a survives today, in the name of a city in India, Kumbhakonam, or Combaconum. The name is very ancient, since it was the capital of the Chola race, one of the oldest of the Hindu dynasties. Oddly enough, Komba is the god of Babinga pygmies, in the heart of French Equatorial Africa. The relationship, if any, between this Aryan-looking word and the Sanscrit kumba, and other Indo-European cognates, is uncertain.

GREEK. Ancient, or Attic Greek makes extensive use of the cognates or derivatives of kumb'a. Most of them carry the meaning of something hollowed out. And so we have: kumbe, meaning the hollow of a vessel, a drinking vessel, cup, bowl; boat; knapsack, wallet. Kumbe also means the head, hence a kind of bird, perhaps the tumbler-pigeon. (Above definitions follow Liddell and Scott's Greek Dictionary).

LATIN. The derivatives of kumb'a, or kumbe are not numerous in classic Latin. I have found only one: cumba (or cymba), meaning skiff boat. In this sense it is famous as the small boat or skiff in which

173

Charon rowed the souls of the dead across the River Styx, in Latin mythology. Listen to Vergil [variant of Virgil], in the Aeneid, Book VI, lines 302-3:

"Ipse ratem conto subigit velisque ministrat et ferruginea subvectat corpora cumba."
(He, himself, ((that is, Charon)) by means of a pole and sails directs the craft, and transports the bodies in a dusky boat.")

And so, it may be said that the squalid ferryman, Charon, rowed the spirits of the departed across the Styx in a COMBS (cumba).

CELTIC. Here we assume an ancient, original, hypothetical kumba, or something very similar to it, meaning (little) valley. The hypothesis is strong, since so many derivatives and closely related words are found in the various Celtic dialects, ancient and modern. They are dispersed over a large part of Europe, as we shall see. To be sure, they do not always mean valley, but they do mean a hollowed out receptacle. If our hypothesis is correct, kumba assumes great importance, since it is in the Celtic tongues that this word (or its later counterparts) begins to designate a valley - the meaning of comb, combe, etc. in later Celtic and English. And what is a valley but something hollowed out, through which a stream flows, or a low land between hills? We must constantly keep in mind the ancient Sanscrit meaning of kumba.

As far as we know, kumb'a did not mean dale, or vale (better translations of comb than valley), although it did mean, usually, something hollowed out. It can easily be imagined that the earliest, crudest skiffs used in crossing streams were large, hollowed out tree trunks, similar to those still used in some parts of the world, and called canoes in the author's part of the country.

1. Gaulish (Gallic). This tongue was practically dead by the fifth century, A.D. Few survivals of its vocabulary are on record. But it has left us cumba, meaning cup, vase, and little barque. Professor Albert Dauzat, in his learned Noms de famille de France, comments on the "gaulois cumba, vallée sèche," in the Midi (South) of France. The meaning of the word sèche (dry) is not clear here. Monsieur Dauzat says further that such diminutives (as family names) as Combette and Comet are found in Gascony. Lacombe is also listed.

2. Breton. This dialect is confined almost exclusively to Brittany today. It is sometimes referred to as Armoric. Its contributions to our study are important, since Breton is one of the few remaining Celtic dialects still spoken. It gives us: komm, a trough; and komb, kombant,

174

koumbant, all meaning a dale, or vale. This is apparently one of the earliest instances of a derivative of kumb'a meaning valley, in any of the Celtic dialects. There is a later, Low Breton comb, meaning little valley.

3. Cornish. Survivals of it are heard today as dialect, in Cornwall, southwestern England. Here we find a more modern cum, meaning dale, dingle. Ewen, in his A Dictionary of Surnames in the British Isles, notes the old Cornish "pen y cum gwic", from whence the modern English corruption, in Devonshire, "Penny-comequick.". It means the head of the little valley, or creek.

4. Irish. The only example which I have discovered here is cumar, a dale; also cum, a cup, vase.

5. Welsh. The present-day representative is cum (or cwm), meaning dale, vale. I shall discuss it later, in place-names. Along with Cornish cum, it looks back toward earlier forms spelled with b. The modern "cwm bychan" means "little combe", or "valley". Cwm is sometimes heard even today, with the sense of a circular geological formation (a sort of basin?). Compare this meaning with that of comb, a cell, in honeycomb. Some scholars in onomastic science have made much of cwm as the origin of Comb(s). Again, it is only the Welsh, or older Celtic form of kumba, comb(e), etc.

PERSIAN. Xumba, a pot, or any hollowed-out receptacle, appears in Zend, an ancient Persian dialect. Modern Persian offers xum(b), a pot, jug. In all probability this word, or both of these words, presuppose an earlier Indo-Iranian kbumbha.

WAKHI. A dialect of Pamir (south central Asia), which submits kubun, a wooden drinking cup. This dialect is not supposed to belong to the Aryan family, but the inhabitants, many of them, live close enough to northern India to bear at least a remote relationship with the Hundus [Hindus?] and their ancient Sanscrit.

THE SLAVONIC GROUP. An examination into the various dialects of the languages of this group would doubtless reveal a large number of words related to or derived from kumb'a. In Russian there are a number of words, meaning crest (hump), and kub, or kubk, and kubok, a cup. Polish reveals kubek, a cup. An entirely new meaning shows up in Czech (Bohemian) kumbálek sent to me by Professor Micek, of the University of Texas. Professor Micek says it means a small room. This is apparently a derivative of kumb'a, and after all, in a sense, a "hollowed out" place. I never heard the word among the Czechs in Czechoslovakia, and assume that it is dialect.

175

THE TEUTONIC GROUP. This is a very extensive group, and, including its dialects, ancient and modern, offers more examples, by far, in the study which we are making, than any other group. A certain amount of "over-lapping" and repetition may seem apparent, after we shall have considered the Romance Group; since much in these two groups is, in its earlier stages, Celtic, which we have already considered. The literary remains of Gothic (the foundation of all Teutonic linguistics) offer no examples for our study. But undoubtedly a number of words listed in this group are of Gothic origin, if I may use the term; which may be another way of saying that they are Celtic? Professor Johann J. Hinrichs, himself formerly from Schleswig-Holstein (bordering Denmark), tells me that he has heard kum (or kom) meaning cup, bucket, etc., among the peasants in that part of Germany; since that is right in the old Gothic world, it smacks of Gothic. But it may be Low German, Frisian, or what will you?

1. Norwegian. Hump, mound, hillock, is the only example I have found here. It is, of course, a survival of our old Sanscrit kumb'a.

2. Dutch. Here I record kom, a bowl, and homp, hump.

3. German. In the various German dialects I shall make use of such abbreviations as: OT (Old Teutonic), OHG (Old High German), MHG (Middle High German), HG (High, or Modern German), and LG (Low German).

OT. Here we find a more or less ancient kumbo, and kummo, meaning a measure, small vessel, tub, cistern, cup, vat.

LG. The representatives are kumb, and kum, meaning a vessel, round, deep basin, trough, bowl, etc. Also kump.

OHG and MHG. The forms are numerous, and I pass on to HG, or Modern German. A number of forms confront us, with the usual meaning of hollowed out receptacles: Coom, kump, kumme, kumpf, kumm, coeme, koome, coome, coumb, cum, kim. One also finds in modern German humpe, a descendant of kumb'a but meaning hump. It is singular, that, as far as my research has carried me, none of the derivatives of kumb'a have come to mean (little) valley, dale, vale or dingle, in either the Slavonic or the Teutonic Group, on the continent. Such a meaning is reserved largely for French and English.

4. English. Here we must consider Old English, Middle English, Early English, and Modern English. It is a long and an intriguing story, the growth and development of the descendants of the parent term kumb'á in the British Isles; for it leads right up, through many

176

stages, to the family name COMBS. (The story is similar in France, Spain and Italy, but certainly less extensive and less complex).

It is amusing to hear certain members of the tribe of COMBS, all over these United States, asserting that the name is Welsh, Irish, or Scottish. Beyond the shadow of any linguistic doubt, it is thoroughly, completely and overwhelmingly English, in its present form; I shall demonstrate this, beyond all cavil.

In Old English cumb (hollow, small valley) occurs in the charters in descriptions of local boundaries, in the south of England. But, for some reason or another, the word (with its varied spellings) does not appear meaning valley in Middle and Early English literature; but this does not argue that it did not have that meaning at the time. There is record of its use (usually that of a hollowed out receptacle, also valley), since the eighth century. In literature it reappears in the second half of the sixteenth century.

Long ago the descendants of kumb'a which meant and mean a small, hollowed receptacle, also small boat, skiff, dropped most of these meanings in the British Isles. But there is an obsolete English combe, or cumb, meaning a hollow vessel, bowl, cup, etc. The Old English spellings seem to be cumb, comb, and camb, and mean small valley, or hollow. The Middle English spellings are mostly coomb, combe, coombe, and coome; and comb, coom, and komb, as in Chaucer and Mandeville. There is also an obsolete (?) combe, etc., meaning a brewing vat, also a measure of four bushels. But I have never heard the word used as such in England.

We are still in the British Isles. At this point, as far as English is concerned, we may as well discard all definitions of comb save that of valley; I mean, the comb descending from Sanscrit kumb'a not those words deriving from Sanscrit "gambhas", a tooth. We cannot determine the exact time at which comb came to mean valley, or hollow in English; but since it appears in Old English documents relating to land boundaries, it must have been early, for the Old English period extended from the fifth century to about the year 1100, roughly speaking. The fifth century was the century of the incursions of the Teutonic tribes (Angles, Saxons and Jutes) into England from the Continent. But it is unlikely that these Anglo-Saxon tribes brought along with them the meaning of valley for comb when they shoved off

177

from the shores of the Continent; for, as stated above, the word never seems to have had that meaning among any of the Continental Teutonic dialects. Since comb is Celtic, it is more than likely that it was in common use in the British Isles long before the invasion of the Anglo-Saxons. Nor does the French word combe (valley), used since early times, owe anything to any Germanic tribe, for similar reasons. As far as the records go, the use of comb for valley may be said to begin with the eighth century; that is, in England. But the usage is doubtless much older than that.

A situation in Low, Medieval or Vulgar Latin worries me: the word cumba or comba appears, meaning valley. (That is identical with the Classic Latin word already mentioned, in Vergil). We also find cumbus, etc., meaning a hollow vessel, basin, trough, bowl, etc. In Classic Latin literature no such meanings seem to exist. It may be that these two words were common in popular or provincial Latin, or that they were borrowed from the Italic dialects, which have much in common with Celtic. This use of cumba for valley is the only one I have found on the Continent, outside France. In the records of the abbey of Sauxillanges (Puy-de-Dôme department, south central France) Cumbas appears as a place-name in the tenth century, for La Combe, a village or community no longer extant. As such, the Latin plural is difficult to explain. Comb means a small valley, dale, dingle, vale, or hollow, in general, all over England; in Wales, cwm (cum), and in Ireland, cumar. Comb is especially common in southwestern England, in Devonshire, and in Cornwall. Very often it means a small valley between two hills, whose stream empties into the sea. It is not common in Ireland and Scotland. A few literary references will illustrate the use of the word as a valley:

"The dark cock bayed above the coomb,
Throned 'mid the wavy fringe of gold."
- [James] Hogg, in Queen's Worke (1813).

Matthew Arnold, in Poems, Youth of Nature: "Far to the south the heath still blows in the Quantock coombs". [Alfred, Lord] Tennyson, in Gareth and Lynette: "Anon they pass a narrow comb.…" [William] Wordsworth, in Excursion: "We dropped with pleasure, into sylvan combs."

It is now time to look at the Romance languages and dialects, and

178

I turn in that direction, where we shall see that valley as a meaning will reappear.

THE ROMANCE GROUP. Here the survivals of kumb'a are also ancient, since, for the most part, they are largely Celtic in origin. With very few exceptions, they mean a valley of some sort. In French and its dialects, that is practically the sole meaning. The French combe meaning a ridge, must not be confused with the same word meaning valley, since, as I have remarked elsewhere, it derives from the Sanscrit gambha, a tooth.

1. Old French. Here a combe is quite common, nearly always meaning a little valley, or a basin, or low place surrounded by hills. Today, one speaks of the combes du Jura, or the vales of the Jura Mountains [boundary extending between France and Switzerland]; and the dwellers in these combes are sometimes called "combiers." In Burgundy [variant of Burgogne region of France] one also hears comme. Combe appears not infrequently in Old and Early French literature, and on down into the eighteenth century. A few references will suffice: Garin le Loheram: "Li os chevauche par tertres et par combes" (The bone rides through hills and vales). Girart de Ross: "Qui estoit â l'iglise assise en une combe" (Who was at the church seated in a crypt). Here the meaning is that of a hollowed out place, or crypt, a rare meaning in French; but compare catacomb (Greek kata, down, plus kumbe, a hollow, cavity, etc.) The French is of course catacombe. [Georges de] Buffon, the great naturalist of the eighteenth century writes: "Dans ces espèces de plaines au-dessous des montagnes, il se trouvent des terrains enfoncés, des vallons secs et froids qu'on appelle des combes." (In these sorts of plains below the mountains there are deep places, little dry and cold valleys which are called combes). It is necessary to make two observations here: vallon means a little valley, and combe may also mean a basin. But Buffon means little valleys, since he uses the modern French vallons. [Jean Jacques] Rousseau, in the same century, says, somewhere: "Dans une combe à vingt pas, j'aperçois une manufacture de bas." (In a valley, twenty paces away, I perceive a hosiery factory).

In Provençal (southeastern France) we find comba, valley. The spelling would indicate Italian dialectic influence, since we find the same word in the Piedmont country, and possibly elsewhere in Italy; but I am not sure. The Low Latin cumba has been mentioned above, and this looks suspicious.

I believe that some form of the Provencial comba is heard among the French Catalonians along the Eastern Pyrenees, but of this I am not certain. Forms of the word appear on the other side of the

179

Pyranees, in Catalonia, northeastern Spain. But we shall have to note the difference in meaning. In the southwestern Alps, in Switzerland, a German Komben, meaning little valleys, comes to light. (The singular form çis Kombe). I have not commented on this form under the Germanic dialects, for the reason that it seems to be purely a borrowing from the French. As far as I know, it has not penetrated into the German mainland.

2. Modern French. In its ascent to modern French, combe has undergone practically no change, either in spelling or meaning. Yet, in modern Provençal comb appears, but not frequently, meaning little valley. The spelling is due doubtless to the influence of standard French, with omission of final e, which is usually pronounced in Provençal. I have already offered some quotations from modern French literature, illustrating the use of combe as a valley. As a place-name it has never made the headway in France that it has in the British Isles. Yet, here and there it appears as such: there is a circular valley in Burgundy called the Combe a la Vielle; also a Combe aux fées (Fairies' Hollow), less than two miles from Dyön; in Franche-Comté there is a place known as the "ferme et Cascades des Combes." The combe de Clavoillon is situated northwest of Beaune, in Burgundy. There is the town of Grand' Combe in the Département of Gard. The contributions of France to the family name will follow later.

Italian. In Italian dialect comba appears; this may be identical with the Piedmontese conba. Both of these may be forms of the Low Latin cumba, already connotating [connoting] a valley; or they may be of Celtic provenance. In the province of Como, in northern Italy, gomba is heard. Maximilien- Paul-Emile] Littre, in his great French dictionary [Dictionnaire de la langue francaise], says that combe is found in a Latin text of the seventh century as a geographical name. He fails to give the source. If this is true (presumably the location is in France), here we have what is possibly the earliest recorded reference to the word as a place-name.

Spanish. Some etymological meandering is necessary in our study of Spanish and its dialects. The meanings of comb, etc. which I have listed and discussed up to this point are totally lacking in the Iberian Peninsula, as far as I have been able to discover. But this does not mean that no [any] derivatives of kumb'a, cumba, etc. do not exist here. In Spanish we find not only nouns as derivatives, but also verbs and adjectives - all of which goes one or two better all the other languages which we have discussed.

In general, comba means, in Spanish, a curve, bend, curvature,

180

convexity; combés means circuit, circumference, or waist of a ship. Combo means bent, or curved; combar, to bend. One sees here a close relationship between all these words and the Greek kumbe, and Latin cumba - by assimilation of the concavity of a barque, cup, bowl, etc. with something bent, curved, concave, etc. In Catalan we find combes, identical with the Spanish combes in meaning. It is probably borrowed from Spanish. Curiously enough, comba also means a children's game of skipping rope among Spanish children. Here again we may see the "curvature" of the rope.

Portuguese. Examples are rare here. I have found only convés, identical with combés.

In the study of place-names we are confined to France and the British Isles. One or two have already been listed for France; there are probably more. But it is the British Isles that concern us here, and England in particular. We have seen that comb(e), etc. as common nouns designating objects, or things, never got themselves firmly fixed in standard English and French, but remained semi-dialectic. But as place-names and family names they become firmly entrenched. It is to this latter consideration that I now turn.

From the prevalence of place-names using comb(e), cum etc. over England and Wales (and in Scotland, to some extent), it is evident that their meaning as valley, or hollow, was extensive, at least among the folk. In Wales and in Scotland the overwhelming preference is Cum. For some inexplicable, linguistic reason, probably in the Celtic age, our Celtic Welshmen and sturdy Scots went back to the ancient u of kumb'a, and dropped the b. And to Cum they stick. The varied spellings of the place names will be illustrated by the examples given. But Cum in place-names occurs also in England, as the following quatrain from an old Cumberland poet will illustrate:

"There's Cumwhitton, Cumwhinton, Cumranton,
Cumrangan, Cumcrew and Cumcatch,
And mony mair Cums i'/ the County,
But nin wi' Cumdivock can match."

But any one can see that this is the work of a Scottish poet, who, after all, makes reference to some county in Scotland, perhaps. Cumberlandshire is in northwestern England, bordering Scotland.

181

When and where does Comb(e) first appear as a place-name? That date cannot be definitely fixed. We stumble upon it occasionally before the Norman Conquest (1066); but here it doubtless often, or usually means a castle, manor house, or estate, named from its location, in a small valley, or at the mouth of a hollow. There were not many cities and towns in the medieval age, and castles, chateaux, manor houses and estates were named.

We learn, from the Cronologica Augustinsis that, more than one thousand years ago Edward Combs (known in history as King Edward the Elder, and son of Alfred the Great), in his "Carta Eduardi de Cumbe," granted land to Gregory. (We shall see that u, and ou, and later oo are sometimes used instead of o). Now, to be sure, this was before surnames become common in England, and the king, Edward, was recording his name, in this particular instance, as "Edward of Cumb", the estate. His reign ended in 924. Later, in the Sarum Charters, we find Gregory de Cumbe, in 990; this is probably the same terra de Cumbe, later known as Combe, located in South Pool, Colerage Hundred, in Devonshire [DEV, EN]. The e in Cumbe, this early in history, before the Norman Conquest (and subsequent French influence), looks suspicious; I opine that the clerk, writing the document in Latin, placed the word Cumb in the third declension, thus adding an e, in the ablative case following the preposition de. U instead of o in the word was not uncommon in Devon, and we shall meet the spelling, or the pronunciation, later, in the family name. The spelling, with u, and ou continues here and there, often and on, for three centuries, and extends outside Devonshire. In Devonshire today one may note Combe Tracey, and the parish of Combe St. Nicholas, in Somersetshire [SOM, EN].

In the same century (the tenth) the records reveal Pyncombe, in Haytor or Colerage Hundred, Devon. Now, Cumb and Pyncombe are without doubt much more ancient than this century; for earlier names of the latter were Estotecoma, Stedcombe, and Combe Pyne. In Devonshire along [alone] there are eighty-two farmsteads, or hamlets called Coombe, in addition to twenty-three others in which Coombe, or some form of it, is used, either with prefix or suffix.

But in Wales the place-names are probably more ancient; since they have undergone less change due to lack of invasions from the Continent, thus preserving their ancient Celtic flavor. Cymro (a Welshman) may be derived from Cum, cwm; a "man of the valley". In the parish ofLlanrotal [Llanrothal] we have the following development of the

182

name of a manor: Cumdu, Cumod, Cumgoy, Cum Mere, Cum Manes, Com Manor (or "The Come"), and finally, in English, Combe Manor. The location is in Herefordshire, now in England, and whose western border lies alongside Wales.

Cum Hir is now an abbey, on the River Ithon, in Wales. Other place-names in Wales are: Cumavon (Glamorganshire), Cumbychan, Cumdauddur (Radnorshire), Cumdare, Cumaman and Cumbach (Glamorganshire), Cum Glas, Cumgors, Cumparce, Cumllynfell, Cum Llinan. The use of the word as a more or less rounded, bowl-shaped hollow or valley inclosed [enclosed] on all sides but one by steep and in some cases perpendicular cliffs (Century Dictionary and Cyclopedia) is also limited to a part of Ireland, especially to county Kerry; where the combs are numerous and of great size, many of them containing lakes. Its old Gaelic or Celtic significance is here manifest.

PREFIXES AND SUFFIXES. Before proceeding with place-names, I remark upon the position of Comb(e), Cum, etc. in the names. We have already seen that in the early English names, it may stand alone as a name. As in the Welsh names, above, it may be used as a prefix, although it is sometimes separated, as in Cum Glas, Cum Llinan, etc. In later English usage it is usually a suffix, such as Pyncombe, Haccombe, etc., etc.; but the rule is not constant. The Celtic Scots seem to prefer it as a prefix, and we have Cumnock, and others. Yet, there is the Scottish White Comb. In Devonshire, alone, it is said that the name appears at least seventy times as a suffix, as in Ilfracombe, Nettlecombe, etc. Compare such modern family names as Whitcomb, Lipscomb, Holcomb, and others.

I must go back a little, to bring up our study. It appears that, after the Norman Conquest, a number of demesnes, or landed estates took the name of Combe. After the Conquest the French e usually is added, but Cumba and Cumbe are also found. It seems that one of the principal demesnes of Edward the Elder, Cumb, passed on down to one of his descendants, King Harold, and that it was this king's seat. After Hastings it went over to William the Conqueror, thence to William fitz Norman de la Mare (son of a prominent Norman nobleman), who was the founder of one of the COMBS families in England. The Conqueror granted Combe Manor to Otho. But the list of Combes over central, southern and southwestern England after the Conquest is extensive, and I shall not discuss them. They are tied up with a number of prominent Combs families. In the Domesday Book (1085-'86) Hundridus de Insula (Humphrey de l'Isle) was lord of Cas-

183

tle Combe. In one of [Walter] Scott's novels, Ralph Avenel forfeits some Combes. Castle Combe, in Wiltshire [EN] was founded by Walter de Dunstanville, and held by Reginald de Dunstanville, known as the Baron of Castle Combe.

Other English place-names: Combemartin, Sadelcombe Parish (Susses), Castle Combe (Wiltshire), Moulds Combe, Combe St. Nicholas, Withycombe, Timbercombe, Combe Hill, Combes Tor, Combe Abbey, and: places held or controlled by Ralf de Pomari de Combe, in the eleventh century: Stedcombe, Selcoma, Vicecomes, Hesmalcombe, Smallacombe, Bichcombe, Combe Temple, Uffcolme, Ilfracombe. In Sherlock Holmes, "The Hound of the Baskervilles," there is Coombe Tracey, a community. To continue: Wallacombe, Cumba, Comba, Combe Fishacre, Ashcombe, Thorncombe, Saccombe, Harcombe, Batancumb, or Batcombe, Brancescumb (Branscombe), Eastcumb (Eastcombe), Sealtcumb (Salcombe), Wincelcumb (Winchcombe), Edgecombe, Luscombe, Lipscombe, Biddescombe, Hanscombe, Compton for Combton ("Combs Town") is a possibility; Welcombe, Woolacombe Sulcombe, Babbacombe Bay, all in Devon; Widcombe and Wiveliscombe, in Somerset.

And so, in order to prove my thesis, I have pulled aside the veil and peered back into four thousand years of linguistic history. I believe that my contention has been established: that COMBS looks back to Sanscrit kumb'a, and to a similar Celtic form, as its prototype; also, that, as far as the present day cognomen or family name is concerned, the meaning is "little valley" or "hollow". Only a pseudo-philologist will claim that it belongs to antiquity as a family name, or even that, in the British Isles, it had any extensive use as such prior to the Norman Conquest. My remarks on family names, in the Introduction, help to bear all this out. Up to this point my study has been of a comparative nature; I shall continue that modus operandi, applying it to genealogical considerations. In addition to the British Isles, I shall invade Italy, France, Spain and Germany.