Translated by Olivier van Renswoude from his original Dutch piece.
Long ago, in a castle in medieval France, there lived a noble couple with their newborn son. One day they had to go out for a short while, leaving him under the guard of their hound Guinefort. When the lady returned the terror was great: the cradle had fallen and Guinefort was stained with blood around his muzzle. Upon her screaming the lord came running in. The hound had devoured their child! The man drew his sword and hewed into Guinefort until his last howl and breath. But it was only afterwards that they found their little son, sleeping safely behind the cradle, and a dead snake nearby, torn to pieces by Guinefort the protector.
With great remorse over this deed they buried him in the well under a large heap of stones and planted trees beside it in his memory. It did not take long before people from all over came to visit the grave and saw Guinefort as a saint, hoping for miracles.
That is the story which was written down in the thirteenth century and can be located near Lyon. It must be much older, however, since it is attested in several forms in a large area. For example, it is practically the same as the Welsh legend of prince Llewelyn and his hound Gelert, and there was a folk tale in India featuring a mongoose as the unlucky protector, which is killed by the woman, rather than the man.
Another version is found in Die seven wijse mannen van Romen (‘the seven wise men from Rome’), a fifteenth century free translation into Dutch of a much older Latin frame story. This seems to ultimately go back to early medieval Persia, but it is unclear whether the story of the faithful hound had already been part of it or whether it was added in Europe. At any rate, the story is of considerable age and captures the essence of the dog, namely man’s best friend. We may keep this in mind when we leave the realm of stories and wrestle with etymology as we go into the nature of his name.
Why the hound is called hound
The word hound, which continues the original Germanic word for ‘dog’, is very old and related to wind in Dutch hazewind and windhond, both meaning ‘greyhound’. Respectively, they go back to Old Germanic *hunda- and *hwinda-, which themselves are reflexes of older, regional Proto-Indo-European *ḱuntó- and *ḱuentó-. In all likelihood they sprouted from the declension of a single noun. That is to say, hound and wind are actually one and the same word. (To be clear: wind in the sense of ‘breeze’ is only a homonym and is of completely different origin.)
Now, this *ḱuentó-/*ḱuntó- has a somewhat strange form. Most other Indo-European languages point to a related but clearly different form as a precursor: *ḱuōn, genitive *ḱunós or *ḱunés. This form was determined on the basis of Greek kúōn, genitive kunós (think of cynology, the study of dogs), Latin canēs (think of canine), Old Irish cú, Lithuanian šuō and Old Indic śván-, among others. Why the (pre-)Germanic form differs from this is a question that we will set aside for the time being.
Etymologists do not agree on the deeper nature of the main form *ḱuōn. On the one hand, the word can be analysed as *ḱu-ōn, i.e. derived from a root *ḱu- with the ending and declension of a so-called n-stem. On the other hand, it can be interpreted as *ḱuōn, as if it were derived with vowel change from a root *ḱuen-. In either case, we are left with the question which root we are dealing with.
The first case
Assuming it was *ḱu-ōn, it has been proposed that this is a corruption of an older *pḱu-ōn and related to *peḱu-, the precursor to Old Germanic *fehu- ‘cattle, wealth’ and hence English fee. Originally, *peḱu- probably referred to sheep, since it appears to have been derived from the root *peḱ- ‘to pluck (wool)’. It would mean that dogs were named after their role in herding. That makes sense semantically, but we would have expected a trace of that *p- in the dog-word in one or more of the daughter languages, yet there is none.
Another proposal is that *ḱu-ōn belongs to a root *ḱeu-, *ḱu- ‘to gleam, shine’, known from Old Indic śóṇa- ‘gleaming, red’ for instance. Dogs would have been named after their coat and literally be known as ‘shiny ones’ or the like. There were and are, however, plenty of animals with shiny coats, so it would not be much of a distinction. Moreover, we may wonder whether the average dog in that time and place was really that shiny to begin with, the poor thing.
It has also been attempted to tie *ḱu-ōn to *h1eḱuo- ‘horse’, which would subsequently have to be read as *h1e-ḱu-o-. A kind of compound then. But since that word is itself of unclear origin, we would not be making much headway.
Lastly, we can try to connect it with the root *ḱues-, which underlies Old Indic śvásiti ‘to sniff, pant, blow, sigh’ and English to wheeze ‘to breathe with difficulty, with a whistle’. It would entail the assumption that *ḱues- is an extention of a simpler root *ḱeu-, *ḱu-, yet there are only faint clues for its existence. The meaning fits well, of course, since audible breathing, be it panting, snorting or sniffing, is typical of dogs.
The second case
But what if the actual word is not to be analysed as *ḱu-ōn, but a plain *ḱuōn? According to linguists like Elisabeth Rieken and Karin Stüber, some of the forms in the daughter language clearly point to this. It would mean we are dealing with a root which was *ḱuen- in its ‘neutral’ form (unstressed also *ḱun-). Such a root did exist, but is easily overlooked since at first glance its apparent meaning is rather hard to reconcile with the nature of dogs.
For we find this root *ḱuen- in the word *ḱuen-to- ‘holy, sacred’, the precursor to Lithuanian šveñtas, Old Church Slavonic svętь and Avestan spənta-. Another derivation seems to have been *ḱun-sló- as the precursor to Old Germanic *hunsla- ‘sacrifice’, attested in forms like Gothic hunsl, Old Norse húsl and archaic English housel ‘Eucharist’. The suffix *-slo- was usually attached to verbs. And indeed, this verb seems to have existed, with Latvian svinêt ‘to celebrate’ as its only known descendant.
Note that the just mentioned derivation *ḱuento- ‘holy, sacred’ has the same form as *ḱuento-/*ḱunto-, the anomalous, pre-Germanic form beside *ḱuōn ‘dog’.
Does this interpretation mean that dogs were, beyond great importance, sacred animals and were revered and celebrated, like brave Guinefort in thirteenth century France? Neither archeology nor tradition indicates that this was the case among the Indo-Europeans. And with regard to Germanic mythology, we can follow Rudolf Simek, an authority on the subject, by saying succintly that dogs barely played a role. For now, the answer seems no.
The nature of dogs
It is much simpler if we reverse the matter. This root *ḱuen-, and with it the verb, can just as well be interpreted as ‘to devote’ or ‘to be devoted to’. It would mean that dogs and hounds were called as they were simply because they are the embodiment of devotion, of complete dedication and loyalty. It is what they are most known for. There is, as it were, a Guinefort in every cur, typified by endless effort and servitude. It only takes a bit of practice.
Derksen, R., Etymological Dictionary of the Baltic Inherited Lexicon (Leiden, 2015)
Kroonen, G., Etymological Dictionary of Proto-Germanic (Leiden, 2013)
Melchert, H.C., “PIE ‘dog’ in Hittite?” in Münchener Studien zur Sprachwissenschaft, Heft 50 (München, 1989), blz. 97-101
Pokorny, J., Indogermanisches etymologisches Wörterbuch (Bern, 1959)
Rieken, E., Untersuchungen zur nominalen Stammbildung des Hethitischen (Wiesbaden, 1999)
Rix, H. e.a., Lexikon der indogermanischen Verben, 2. Auflage (Wiesbaden, 2001)
Simek, R., Lexikon der germanischen Mythologie, 3. Auflage (Stuttgart, 2006)
Stüber, K., The Historical Morphology of N-stems in Celtic (Maynooth, 1998)
Wodtko, D. e.a., Nomina im Indogermanischen Lexikon (Heidelberg, 2008)