[PLUG-TALK] dog?

John Jason Jordan johnxj at gmx.com
Tue Mar 27 18:15:15 PDT 2018

On Mon, 26 Mar 2018 22:13:31 -0700
Galen Seitz <galens at seitzassoc.com> dijo:

>Sending this to the list for others amusement, but hoping John can

Historical Linguistics
(Partly recycled from a series of short discourses given at PSU)

Prior to the 1800s the field of linguistics didn't exist, and even
through that century it was concerned with just philology, i.e., word
origins. Of course, at that time and for centuries before scholars had
been required to study Latin to the point of fluency because university
clases were taught in Latin. Ancient Greek was also popular with
scholars. It was obvious to just about everyone that there were
connections between Ancient Greek and Latin, not just for their
lexicons but also for morphosyntax. Just to give you a sample of the
morphosyntax, nouns carried endings (inflections) denoting their
function in a sentence and verbs were even more heavily inflected. Here
are some of the more common case inflections for Latin canis 'dog' and
Ancient Greek κύων, κυνός, ὁ, ἡ 'dog' (brackets = IPA, note that in
IPA : = long vowel and y = ü, as in German):

nominative (subject)	canis	[kanis]	κύων [kyo:n]
genitive (possessive)	canis [kanis]	κυνός, [kynos]
dative (to or for)		canī [kani:]	κύνι [kyni:]
accusative (object)	canem [kanem]	κυνα [kyna]
Plural (dual forms omitted for brevity)
nominative (subject)	canēs	[kane:s]	κύνες [kynes]
genitive (possessive)	canum [kanum]	κυνων [kyno:n]
dative (to or for)		canibus [kanibus]	κύνσι [kynsi:] 
accusative (object)	canes [kanes]	κυνας [kynas]

Contrast these with Proto-Indo-European (PIE) forms *[kolignos] and
*[kuo:], believed to have been dialect alternations. (In linguistics
'proto-' = reconstructed from evidence found in daughter languages.) In
Indo-European (IE) the *[kuo:] form would have added [nt] to the stem
before adding the inflections, thus for most forms the stem would have
been *[ku:ont-]. (Linguists use * to mark reconstructed forms for which
there is no direct evidence.)

At this point I must suggest that your understanding of what follows
here will benefit greatly from reading Wikipedia on

For centuries scholars had noted the similarities among the lexicons
and morphosyntaxes of what we now know as the Indo-European language
family, but it was not until 1786 when William Jones wrote a definitive
paper on the subject that the world of philology changed from merely
working on isolated word origins to what later became known as
historical linguistics. An Anglo-Welsh philologist, Jones worked in
India where he studied Sanskrit and immediately realized its
similarities to Latin, Ancient Greek, and even English. To this day
Indo-Europeanists work mostly with Latin, Ancient Greek and Sanskrit,
being the closest to the original IE. And in case you declined to read
the Wikipedia article, linguists postulate that IE was spoken somewhere
around the Black Sea roughly 5500-6000 years ago. The exact location
and date are vehemently argued in the field. We know that the
Indo-Europeans really got around; daughter languages have been found
all over Europe and as far east as China (Tocharian). (The earliest
archaeological evidence of the the wheel is from ~10,000 years ago; add
the invention of the cart a few thousand years later, plus the
domestication of the horse, and herding and agriculture, make it clear
why the Indo-Europeans were able to take over lands occupied by
hunter-gatherers.) And because the Europeans later spread their
languages all over the Americas as well as to places on other
continents, today nearly half the world's population speaks an IE

And now, as you may have already guessed, I am going to point out how
IE [ku:ont-] is the origin of English 'hound,' although I have to start
with some basics.

In the late 1800s a group of European philologists known as the
neogrammarians came up with the idea that as sounds changed in a
language the sound changes were invariably regular. From Wikipedia
above: "... compare the pairs of words in Italian and English: piede
and foot, padre and father, pesce and fish ... there is a consistent
correspondence of the initial consonants that emerges far too
frequently to be coincidental." Now, you may be wondering how a [p]
could become an [f] or vice-versa. To understand that we must take a
side trip into the fields of phonetics and phonology.

All human languages have an inventory of sounds, and no two languages
have exactly the same inventory; indeed, the main distinction between
dialects is that their sound inventories differ slightly. Phoneticians
and phonologists describe sounds mainly by (1) the place of
articulation in the mouth, (2) the manner of articulation and (3)

Places of articulation, with an English example: 
	bilabial [p], labiodental [f], dental [θ], alveolar [t], palatal [j], velar [k], plus
	a few more places without English examples. 
Manner of articulation, with an English example: 
	stop [p], fricative [f], nasal stop [n], plus a few more. 
Voiced vs. voiceless, with English examples of voiceless/voiced: 
	[p,b], [f,v], [θ,ð].

Sounds in a language can change in all kinds of ways, but it is unusual
for a sound to change in more than one way at a time, e.g., a [p] might
become a [b] (voicing), or it might become a [ɸ] (fricativization), or
in might become an [m] (nasalization), but in becoming each of those
sounds it retained its place of articulation (bilabial). Or a sound
might move to an adjacent place of articulation, but in so doing it
would retain its manner of articulation and voicing. If we observe a
sound change that involves two changes, then we can be pretty sure that
there were really two sound changes over historical times. Referring to
the Wikipedia quote above, we know from our study of IE that the Italian
forms 'piede, padre and pesce' were older than the English 'foot,
father and fish.' In fact, the change from [p] to [f] in the Germanic
languages is documented in historic times, cf. English 'pepper' and
German 'Pfeffer.' 

All sound changes are the result of either lenition or fortition
(weakening or strengthening), most commonly lenition, that is, humans
are basically lazy so speakers seek easier ways to pronounce things. As
a result, while fortition does occur, lenition is massively more
common. Now, a fricative is just a stop, e.g., [p], where you don't
completely close the opening, resulting in turbulence instead of a
plosive sound. Not completely closing the opening is easier than
closing it, so stops quite commonly become fricatives, e.g., [p] > [pf]
> [f]. 

I promised that I would show how Latin canis and Ancient Greek 'κύων'
are the same as English 'hound.' Well, if [p] > [pf] > [f] is possible,
why couldn't [k] become [kh] and then [h]? Indeed, that is exactly what
happened. And again, we know the change from [k] to [h] was relatively
recent in the Germanic languages, and we have lots of other examples -
Latin 'cornus' = English 'horn,' Latin 'collis' = English 'hill,' and I
could go on and on. So, given the [k] > [h] change, IE [kuont-] could
easily become English 'hound.' In fact, it is surprising that there
were not more changes to [ku:ont-] in the past 5500-6000 years.

From the Oxford English Dictionary: 'hound'
"Common Germanic: Old English hund = Old Frisian hund, hond, Old Saxon
hund (Low German hund, Middle Dutch hont (d-), Dutch hond), Old High
German hunt (d-), (Middle High German hunt, German hund), Old Norse
hundr (Swedish, Danish hund), Gothic hunds < Old Germanic *hundo-z,
generally held to be a derivative of base *hun-, pre-Germanic *kun-, in
Greek κύων, κυν-, Sanskrit çwan-, çun-, Lithuanian szů, szun-, Old
Irish cu dog; compare also Latin canis. For the d (dh) of Germanic
hund, the suggestion has been made of association with the verb hinþan
to seize, as if the word were understood to mean ‘the seizer’." My
note: I have always wondered if English 'hunt' was related to 'hound.'
If the OED agrees with my theory, they didn't say so, either here or in
their etymology of 'hunt.'

So what of 'dog.' Well, that is harder to explain because we don't know
exactly where it came from: 
From the Oxford English Dictionary: 'dog'
"Origin unknown. Earliest attribution in Anglo Saxon: 1175 CE
The word belongs to a set of words of uncertain or phonologically
problematic etymology with a stem-final geminated g in Old English
which is not due to West Germanic consonant gemination and therefore
does not undergo assibilation. These words form both a morphological
and a semantic group, as they are usually Old English weak masculine
nouns and denote animals; compare frog n.1, hog n.1, pig n.1, stag n.1,
Old English sugga (see haysugge n.), Old English wicga (see earwig n.),
and perhaps teg n.1 It has been suggested that these words show
expressive gemination, perhaps due to their being originally
hypocoristic forms. (For discussion see R. M. Hogg ‘Two Geminate
Consonants in Old English’ in J. Anderson Lang. Form & Ling. Variation
(1982) 187–202.) For some of the words, substratal influence has also
been considered (compare pig n.1). Because attestation of these words
in Old English is generally rare and confined to glossaries and
onomastic evidence (as in the case of dog n.1), if they are attested at
all, and also because there is often a better-attested synonym (in this
case, hound n.1), it seems likely that the words were stylistically
marked in Old English, i.e. considered non-literary or informal."

But I must insert here another issue of language change: You need some
knowledge of linguistics to follow and understand sound changes, but
lexical changes are far easier. Languages borrow words and terms from
each other at the drop of a sombrero. Today English is the world's
major donor language, but in past centuries that was not at all the
case. For a couple centuries following 1066 English borrowed massively
from Norman French, and still borrows from French today; around 18-28%
of the English lexicon, depending on how you count. And English has
always borrowed from Latin and Greek, especially Latin. Of course,
English really no longer borrows from Latin - we can't, we've already
borrowed the whole lexicon. We've even borrowed Latin bound morphemes
like -ity, -itude, -tion (-sion), -ine, and many more. So, since we are
speaking of dogs and hounds, how about some other English terms for
these beloved animals? Here are some excepts from the OED:

"Origin: Probably formed within English, by derivation. Etymons: mung
n.1, mang v.1, -rel suffix. Etymology: Probably < mung n.1 or mang v.1
+ -rel suffix. With form mengrell perhaps compare forms in -e- s.v.
ming v.1"

"Etymology: Middle English curre corresponds to Middle Dutch corre
‘canis villaticus, domesticus’ (Kilian), Swedish and Norwegian
(widely-spread) dialect kurre, korre ‘dog’, etc. The latter is
generally associated with the onomatopoeic verb Old Norse kurra to
murmur, grumble, Swedish kurra to grumble, rumble, snarl, Danish kurre
to coo, German obsolete and dialect kurren to growl, grumble, murmur,
coo, compare gurren to coo, Middle High German gürren to bray as an
ass. The primary sense appears thus to have been ‘growling or snarling
beast’. But no corresponding verb appears in English, so that Middle
English kurre was probably introduced from some continental source. The
combination kur-dogge is met with considerably earlier than the simple
kurre, cur. Senses 2, 3 are possibly independent echoic formations."

"Etymology: Old English hwelp = Old Saxon hwelp, (Middle) Low German,
(Middle) Dutch welp, Old High German (h)welf (Middle High German,
German welf), Old Norse hvelpr (Swedish valp, Danish hvalp): further
relations uncertain."

"Etymology: Apparently < Middle French, French poupée poupée n.
(compare -y suffix5); compare poppet n. and puppet n., which are
probably ultimately related to this word, and which show some semantic
overlap; compare also pup n.1, which is probably shortened from this
word, although the earliest examples of each word are somewhat
ambiguous (see discussion below). Sense 1 is not paralleled in French,
and is shown rather later by puppet n. (see puppet n. 9), perhaps under
the influence of the present word; this sense perhaps developed (in
spite of the chronology) from sense 2 in English, or perhaps from the
sense ‘childish person’ in French, or perhaps from the identification
of both a doll and a pet dog as ‘playthings’ (however, the transferred
sense ‘hobby’ in French is first attested much later (1694), and seems
unlikely to play a part in the explanation). The spelling popis in
quot. 1486 at sense 1a could instead be interpreted as showing pup n.1,
with -is showing the plural ending (as also could poppys and popys in
manuscripts from the same textual tradition, in which -ys is the
regular plural ending, but in which for example connys can also be
found for the plural of coney n.1)."

And while I am discussing the lexicon, I would be remiss if I didn't
point out that we don't just borrow words, we change the meaningsof
existing words as well. You have already seen some changes if you
read the OED excepts above. Changes in meaning are usually slow. Here's
an example from contemporary English: A couple years ago I overheard a
young lady say to a friend "My brother backed his car into a telephone
pole! He's amazing!" Now, I could have called the brother a lot of
things, but not 'amazing.' In my English 'amazing' had either a neutral
or positive marker on it, never negative. Apparently the meaning of
'amazing' has undergone a slight change. But here is a famous example
(that not all linguists buy, including me): There was a word in an IE
language 'belk' that meant 'light from a fire, or in some contexts
'white.' Over time it came to mean the charred remains of the fire. And
eventually it came to mean the color of the charred remains, i.e.,
black. Thus, white came to mean black and, to take the theory to its
end, English 'black' and Spanish 'blanco' (white) are cognates.

At this point those of you who like programming languages are probably
glad you are programmers.

OK, that little discourse was lots of fun for me, but now I must get
back to real work. :(

More information about the PLUG-talk mailing list