Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

OE assistance please: How do I properly pronounce "hw?te"

4 views
Skip to first unread message

WJW

unread,
Jan 11, 2008, 11:22:46 PM1/11/08
to
Hello from Colorado Springs,

I am hoping someone may be willing to share their Old English Saxon
expertise by providing me with some Modern English word sounds that may lead
me to a more accurate 6/7th century pronunciation of the OE word "hw?te"
meaning "wheat". If possible, I would also very much like to attain the OE
Saxon word conjunction for either "wheat growing in a meadow" or "wild
wheat".

Last but not least, I am extremely interested in tracing the origin of the
word "hw?te" in terms of its Germanic mainland parental source. Because
wheat is a "founding crop", I believe it is an extremely ancient word . .
.but where did the Germanic and Old Norse versions originate from?

Thank you very much for any assistance that me be forthcoming

Respectfully,

William
qua...@comcast.net

"My dog is so smart he can smell dinner 30 minutes before its made!"


WJW

unread,
Jan 11, 2008, 11:26:22 PM1/11/08
to
Please excuse the "?" in my above post as I just realized the post did not
carry the proper code. I am in fact referring to the word "hwaete".

William


Peter T. Daniels

unread,
Jan 12, 2008, 10:06:35 AM1/12/08
to

"hw" is the same as modern "wh" -- if you pronounce "which" and
"witch"differently, you know how to do it (but most English-speakers
don't make that distinction any more). "ae" is like the "a" in "flat",
and you have to say the -e at the end, which is a little nothing of a
vowel. So "whaete" will rhyme with a British pronunciation of
"flatter" (without the r). The "ae" is marked with a macron, it's a
long ae.)

According to Watkins's *American Heritage Dictionary of Indo-European
Roots*, it's IE *kweid- 'white' in the suffixed o-grade form *kwoid-
yo-.

According to Carl Darling Buck's *Dictionary of Selected Synonyms in
the Principal Indo-European Languages*, this word for 'white' appears
in Germanic and Indo-Aryan (other IE families have different words for
'white'); and only Germanic uses 'white' as the basis of the word for
'wheat'. (This suggests that wheat wasn't known to the original
speakers of Indo-European, but as it was introduced to speakers of
different IE language families, it got named differently -- sometimes
as specialization of a general word for 'grain', indicating its staple
condition.)

WJW

unread,
Jan 12, 2008, 11:19:30 AM1/12/08
to
Thank you Peter . . . that is a big help!

In essence, if the equivalent Modern English word sound is akin to "whatte",
than would you concur that a 6th century Saxon would likely describe "wheat
growing in a meadow" as "hwaetele" wherein the pronunciation would be
"whatte-ley"?

Given the lack of written Saxon documents and maps in Britain predating the
Norman Conquest, I guess it is not surprising that I can not find any Saxon
use of the word "hwaete" in the UK National Archives (macron implied).
However, the word does appear in a number of documents/maps at and after AD
1066 with the "hw", "wh" and "wa" spellings. So . . . at what point did
the spelling begin to change from "hw to wh" and do you think this change is
indicative of an attempt to attain a stronger British identity for the
developing language?

The reason I ask is that the French-Norman language appears to have favored
the Vulgar Latin words for wheat rather than the Germanic hwaete or the Old
Norse hvete (at least in Normandy). Is it therefore possible that the
spelling change from "hw to wh" is the result of an interim Anglo-Norman
language?

William


"Peter T. Daniels" <gram...@verizon.net> wrote in message
news:54241985-1926-4cb5...@v29g2000hsf.googlegroups.com...

Trond Engen

unread,
Jan 12, 2008, 1:09:15 PM1/12/08
to
Peter T. Daniels skreiv:

> On Jan 11, 11:26 pm, "WJW" <qua...@comcast.net> wrote:
>
>> Please excuse the "?" in my above post as I just realized the post
>> did not carry the proper code. I am in fact referring to the word
>>> "hwaete".
>

> [...]


>
> According to Watkins's *American Heritage Dictionary of Indo-European
> Roots*, it's IE *kweid- 'white' in the suffixed o-grade form *kwoid-
> yo-.

(Following and paraphrasing Bjorvand and Lindeman:) A palatal k (/k^/)
since it's found in satem languages with (some sort of) s. Into Germanic
as *hWait-ija-, in West Germanic also with j-gemination *hWait-ja- >
*hWaitt-ja-, apparently. (I'm using j where anglophones often prefer y.)

> According to Carl Darling Buck's *Dictionary of Selected Synonyms in
> the Principal Indo-European Languages*, this word for 'white' appears
> in Germanic and Indo-Aryan (other IE families have different words
> for 'white');

There are cognates in Slavic, although not in the meaning "white", e.g.
OCS <světŭ> "light".

> and only Germanic uses 'white' as the basis of the word for 'wheat'.

B&L quote Lithuanian <kvietỹs> as a Germanic loan. Obviously so, as
Lithuanian would have šv- in an inherited word. I think the initial kv-
dates the loan as either very old (Pre-Germanic) or fairly recent from
some dialect with *hW- > /kv-/. Nowadays that feature is limited to West
Scandinavian (including(?) Shetland) and some subgroup of Finland
Swedish, and, since I'm not aware of a wider distribution historically,
I'll go for Pre-Germanic. I suppose that the long y supports the
reconstruction of a (Pre-)Germanic *-ija.

--
Trond Engen
- sort of corny

Trond Engen

unread,
Jan 12, 2008, 1:13:27 PM1/12/08
to
Trond Engen skreiv:

> I suppose that the long y supports the reconstruction of a
> (Pre-)Germanic *-ija.

It might, but it's nasal, not long.

--
Trond Engen
- nasal, not long, himself, these days

Trond Engen

unread,
Jan 12, 2008, 1:38:57 PM1/12/08
to
WJW skreiv:

> Thank you Peter . . . that is a big help!
>
> In essence, if the equivalent Modern English word sound is akin to
> "whatte",

The length of the /æ/ is important. It's as long as the vowels of in
'weed' and 'fart'.

> than would you concur that a 6th century Saxon would likely
> describe "wheat growing in a meadow" as "hwaetele" wherein the
> pronunciation would be "whatte-ley"?

Long /æ/. And it's the first part of a compound that modifies the
second. An OE *hwaete-lee would not be "the 'lee' type of 'hwaete'" but
"the 'hwaete' type of 'lee'", i.e. "wheat-field".

> [...] So . . . at what point did the spelling begin to change from


> "hw to wh" and do you think this change is indicative of an attempt
> to attain a stronger British identity for the developing language?

As opposed to the utterly French habit of starting words with hw-? No.

> The reason I ask is that the French-Norman language appears to have
> favored the Vulgar Latin words for wheat rather than the Germanic
> hwaete or the Old Norse hvete (at least in Normandy).

I don't really get your idea. Norman is a dialect of French and a
descendant of the local variety of Vulgar Latin. No Anglo-Saxon involved
at all. There are no more than a handful of loans from 9th century
Scandinavian settlers, so there's no reason why the inherited word for
wheat should be Germanic. If the local word for wheat were anything
other than Romance it should be Celtic.

> Is it therefore possible that the spelling change from "hw to wh" is
> the result of an interim Anglo-Norman language?

No. I think it may reflect a weakening of the aspiration, since it's no
longer percieved as precluding the /w/, but such weakening is a
development shared with West Germanic and Southern Scandinavian. It may
also be just a new ortographic style.

--
Trond Engen
- sifting carefully

Peter T. Daniels

unread,
Jan 12, 2008, 1:47:49 PM1/12/08
to
On Jan 12, 11:19 am, "WJW" <qua...@comcast.net> wrote:
> Thank you Peter . . . that is a big help!
>
> In essence, if the equivalent Modern English word sound is akin to "whatte",
> than would you concur that a 6th century Saxon would likely describe "wheat
> growing in a meadow" as "hwaetele" wherein the pronunciation would be
> "whatte-ley"?

I have no opinion whatsoever on matters concerning Old English.

> Given the lack of written Saxon documents and maps in Britain predating the
> Norman Conquest, I guess it is not surprising that I can not find any Saxon
> use of the word "hwaete" in the UK National Archives (macron implied).
> However, the word does  appear in a number of documents/maps at and after AD
> 1066 with the "hw", "wh" and  "wa" spellings.   So . . . at what point did
> the spelling begin to change from "hw to wh" and do you think this change is
> indicative of an attempt to attain a stronger British identity for the
> developing language?

The OED records changes in spelling, with dates. _That_, you can
easily look up for yourself!

> The reason I ask is that the French-Norman language appears to have favored
> the Vulgar Latin words for wheat rather than the Germanic hwaete or the Old
> Norse hvete (at least in Normandy).  Is it therefore possible that the
> spelling change from "hw to wh" is the result of an interim Anglo-Norman
> language?
>
> William
>

> "Peter T. Daniels" <gramma...@verizon.net> wrote in messagenews:54241985-1926-4cb5...@v29g2000hsf.googlegroups.com...

Brian M. Scott

unread,
Jan 12, 2008, 2:59:11 PM1/12/08
to
On Sat, 12 Jan 2008 09:19:30 -0700, WJW <qua...@comcast.net>
wrote in <news:H9ydnbQ3NMjcdRXa...@comcast.com>
in sci.lang:

> Thank you Peter . . . that is a big help!

> In essence, if the equivalent Modern English word sound is
> akin to "whatte",

Don't forget, though, that the first vowel is long.

> than would you concur that a 6th century Saxon would
> likely describe "wheat growing in a meadow" as
> "hwaetele" wherein the pronunciation would be
> "whatte-ley"?

No. The compound that you're trying to make is <hwæteleah>
in the nominative case, where both the <æ> and the diphthong
<ea> are long. The latter is pronounced roughly [æ@], where
[æ] is the <cat> vowel, and [@] is the schwa vowel spelled
<a> in <sofa> and <about>. The <h> is pronounced [x]
(informally \kh\), like German <ch> in <Bach>.

However, <leah> (<læh> in Anglian dialects, with long <æ>)
did not originally mean 'meadow', but rather 'woodland
clearing, glade, underwood'. From about the 8th century on
it was the usual term for woodland clearings, succeeding
<w(e)ald>.

> Given the lack of written Saxon documents and maps in
> Britain predating the Norman Conquest, I guess it is not
> surprising that I can not find any Saxon use of the word
> "hwaete" in the UK National Archives (macron implied).

Annal 1039 of the Anglo-Saxon Chronicle (manuscript E, MS
Laud 636, known as the Peterborough Chronicle), speaks of
the price of <se sæster hwætes> 'the sester [a measure] of
wheat'.

> However, the word does appear in a number of
> documents/maps at and after AD 1066 with the "hw", "wh"
> and "wa" spellings. So . . . at what point did the
> spelling begin to change from "hw to wh"

The early Anglo-Normans generally wrote plain <w> for the
sound. Just scanning a few personal and place-name
references, I see occasional instances of <wh> for this
sound in the 12th century. But there are occasional
pre-Conquest instances; the C manuscript of the ASC has
<while> for <hwile> in annal 1045, and the A (Parker) ms. hs
<whenne> in annal 1031.

> and do you think this change is indicative of an attempt
> to attain a stronger British identity for the developing
> language?

No.

> The reason I ask is that the French-Norman language
> appears to have favored the Vulgar Latin words for wheat
> rather than the Germanic hwaete or the Old Norse hvete

That's <hveiti>.

> (at least in Normandy).

???

Old French, including Anglo-Norman, was descended from
Vulgar Latin, but it did not in general use VL words as
such. Official documents of that period are generally in
Latin, but that's another matter altogether. (And <blet>,
the Old French 'wheat' word that survives in modern French
as <blé>, was a borrowing from Frankish or Gaulish, not an
inheritance from VL.)

> Is it therefore possible that the spelling change from "hw
> to wh" is the result of an interim Anglo-Norman language?

Not in the way that you seem to have in mind. It's true
that some Continental orthographic practices were taken over
into English spelling after the Conquest; an obvious example
is the use of <ch> for the sound in <church>, a sound that
in Old English was spelled simply <c>. Similarly, <th>
displaced <þ> (thorn) and <ð> (edh). Our <sh> sound was
spelled <sc> in OE; the early Anglo-Normans didn't have the
sound, and early spellings are all over everywhere -- first
<s> and <ss> alongside the traditional <sc>, later <s>,
<sch>, <ssch>, <ssh>, and <sh>, with <sh> finally
establishing itself as the standard. By the mid-13th
century or so there was a general pattern of <h> being used
as a diacritic in digraphs to indicate a modification of the
sound of the first letter of the digraph, and that tendency
undoubtedly favored the use of <wh> at the expense of older
<hw>. (In the north <qu> was favored, however.)

Brian

Dušan Vukotić

unread,
Jan 12, 2008, 3:58:13 PM1/12/08
to
On Jan 12, 5:19 pm, "WJW" <qua...@comcast.net> wrote:
> Thank you Peter . . . that is a big help!
>
> In essence, if the equivalent Modern English word sound is akin to "whatte",
> than would you concur that a 6th century Saxon would likely describe "wheat
> growing in a meadow" as "hwaetele" wherein the pronunciation would be
> "whatte-ley"?

This is a very interesting discussion!
Let me compare the Serbian word 'pšenica' and English 'wheat'. Would
anyone have ever supposed that these two words appeared from tha same
Ur-basis? Probably not! And the man who would proposed something like
that (as I intend to do right now) would probably be named as a
complete ignoramus or a nut case. Considering the fact that I am just
about to make such a proposal, what do you think am I a crackpot or
not? ;-)

Namely, I am not 100% sure is English oats is related to wheat, but
comparing it with ON hafri, Latin avena (oats) and Serbian ovas (oats)
both these words (wheat and oat) should be offspring of the same
"parent". Similar is with the Serbian words 'pšenica' (wheat) and
'ovas' (oat). As I underline many times before, there were no IE word
which could originally begin with a wovel as an initial sound. It
means that 'ovas' and 'avena' firstly sounded as 'hovas' and
'havena'.

The biggest surprise is still to come. All the above-mentioned words
have nothing to do with some of the inherent characteristics of wheat
or oat but are related to water! Formal comparsion of English wheat/
oat with word wet should be one of the serious reasons for a deeper
investigation of my above statement.

One of the main Serbian meals for centuries was a meal called "ovsena
kaša" (oat soup). That oat meal was also called 'kvašenica' ('kaša' /
soup/ is a shortened form of 'kvašenica') because it was prepared of
no other ingredients instead of oat flour and water (eventually
salted). That 'kvašenica' (ovsena kaša) could be translated to English
as "wetted oat".

At the same time, this 'kvašenica' was also named as
'ovsenica' (ovsena kaša again or "wet oat") and by applying the
Saussure's Theory in the sense that its author imagined it (and not in
sense of Cyba-Cave gurus) we are getting the word 'hovsenica'; i.e.
above 'kvašenica' (made of ovas or hovas /oats/).

Finally, I must add that 'kvašenica' is derived from the noun
'kvašenje' (soak, suffusion) and verbs 'kvasiti', ukvasiti', which are
then related to Latin aqua and aquaticus.


DV

WJW

unread,
Jan 12, 2008, 4:06:45 PM1/12/08
to
Thank you all for the excellent insight provided! As an archaeologist, my
interest is strong but my etymological knowledge and skills are very
limited. As such, I greatly appreciate the response.

My surname is "Whatley", hence the reason for my search. We know from
family documents and oral history that the family came to Wessex from
Normandy during the Norman Conquest. The name appears in Somerset as
"Watelei" and as "Watelege". However, I have had little success in
confirming similar use in Normandy. This baffles me somewhat because our
family history says that prior to Normandy, we came from Norway where we
reportedly still have distant relatives using the name "hveitile" or
"hvetileah". For what its worth, our oral history says the name was
initially given as a Norse descriptor for long, uneven blonde hair that
looked like wild wheat.

Thanks again!

Admittedly out of my league, I am, respectfully yours,

William


Dušan Vukotić

unread,
Jan 12, 2008, 4:15:36 PM1/12/08
to

Sorry, I forgot to say that Serbian word 'pšenica' came from
'hovsenica' or 'ovsenica/obsenica' => pšenica.

So the final conclusion might be:

Eng. wheat => oat
Serb. ovas (oats) <= ovsenica/kvašenica => pšenica (wheat)

I wonder if there was any meal made of wheat or oat that was named
close to "wet oat", "quas-' or something similar?

DV

Message has been deleted

Brian M. Scott

unread,
Jan 12, 2008, 5:21:18 PM1/12/08
to
On Sat, 12 Jan 2008 12:58:13 -0800 (PST), Dušan Vukotić
<dusan....@gmail.com> wrote in
<news:ce316eb0-e5e3-4849...@u10g2000prn.googlegroups.com>
in sci.lang:

[...]

> Namely, I am not 100% sure is English oats is related to wheat,

Not.

> but comparing it with ON hafri,

The word is found only in the plural, <hafrar>, and it is
not found in early writers.

[...]

Trond Engen

unread,
Jan 12, 2008, 7:23:17 PM1/12/08
to
WJW skreiv:

> Thank you all for the excellent insight provided! As an
> archaeologist, my interest is strong but my etymological knowledge
> and skills are very limited. As such, I greatly appreciate the
> response.
>
> My surname is "Whatley", hence the reason for my search. We know
> from family documents and oral history that the family came to Wessex
> from Normandy during the Norman Conquest. The name appears in
> Somerset as "Watelei" and as "Watelege".

The name looks unmistakenly English and a quick web search tells me that
Whatley is the name of a village in Somerset
(<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are Norman
they certainly didn't bring that name with them.

The name is supposed to mean "wet-glade". That would mean that the h is
unetymological. The "wate-leah" would have been the name the locals used
for that particular opening in the forest and the name they continued to
use after it was settled. Looking at maps* I see that the village is
situated on a low hill between two brooks. What's more, there's a
vineyard at the manor, AFAIK not a particularly common sight on wet
soil. I'd suggest that the "water" part is refering to one or both of
the brooks, rather than to the wetness of the soil.

Another option, if wh- is etymological, is that the first part
corresponds to ON *hvatr "quick". It's an element in several farm names
in Norway and is thought to describe "(soil) with good growth"
(<http://www.dokpro.uio.no/perl/navnegransking/rygh_ng/rygh_visetekst.prl?s=n&Vise=Vise&KRYSS112251%4025515=on>).
It seems to fit well with the denseness of fields (and the vineyard) on
the map, but I'm uneasy about using a description of the soil or growth
in connection with the element -leah, since there was no farming there
when it was a leah. Unless, of course, 'leah' had come to mean "new
farm" at the time.

There's also a Norwegian farm Hvaterud < *Hvataruð whose first name
element as an alternative is suggested to be <Hvati>, a byname for names
ending in the element <-hvatr>. This is highly unlikely in a Somerset
toponym, but I mention it since I think the suggestion is due to a
similar objection to the compound.

Admittedly, though, I know nothing about placenames in Somerset.

> However, I have had little success in confirming similar use in
> Normandy. This baffles me somewhat because our family history says
> that prior to Normandy, we came from Norway where we reportedly still
> have distant relatives using the name "hveitile" or "hvetileah".

No, you have not! The Scandinavian settlement in Normandy took place in
the late 9th or early tenth century. (The duchy of Normandy was founded
by a peace treaty in 911, but I think most of the settlement happened
earlier, making the treaty a confirmation of a fact.)

First, Scandinavians didn't have inherited names at the time, or for
centuries to follow, so no family name can have survived. Neither had
the English, by the way.

Second, there are only some five or ten family lines in Norway that can
be traced to the 14th century and, if memory serves me, none except the
royal family before the 13th (and that line has some weak links, too).

Third, I think it's generally believed that most (or all) of the few
settlers were Danish. That's utterly uncertain, though. The settlement
happened before the political map of Scandinavia was drawn. Everybody
spoke 'danska' (Danish) and 'norðmaðr' (Norman) simply meant "man from
north".

> For what its worth, our oral history says the name was initially
> given as a Norse descriptor for long, uneven blonde hair that looked
> like wild wheat.

*hveitiligr? "wheat-like". No, they'd come up with something more
inventive, like "sheaf".

> Admittedly out of my league, I am, respectfully yours,

*Dozens of map links for the Whatley Quarry are provided in the
Wikipedia article. Here are three:
Ordnance Survey:
<http://getamap.ordnancesurvey.co.uk/getamap/frames.htm?mapAction=gaz&gazName=g&gazString=ST7314947950>
Google Maps:
<http://maps.google.com/maps?ll=51.229591,-2.384583&spn=0.025,0.025&q=51.229591,-2.384583>
MSN UK:
<http://www.streetmap.co.uk/newmap.srf?x=373150&y=147950&z=3&sv=373150,147950&st=4&mapp=newmap.srf&searchp=newsearch.srf&ax=373150&ay=147950>


--
Trond Engen
- Lee? What lee?

Brian M. Scott

unread,
Jan 12, 2008, 7:53:31 PM1/12/08
to
On Sat, 12 Jan 2008 14:06:45 -0700, WJW <qua...@comcast.net>
wrote in <news:F4CdnUo3PJ4JthTa...@comcast.com>
in sci.lang:

[...]

> My surname is "Whatley", hence the reason for my search.
> We know from family documents and oral history that the
> family came to Wessex from Normandy during the Norman
> Conquest. The name appears in Somerset as "Watelei" and
> as "Watelege".

The name is undoubtedly locative in origin, from an English
place-name. The most probable source is Whatley in
Somerset. This name is in record as <Watelea> 939x946 (in a
13th century copy), <Watelei> 1086, <Wateleg> 1225, and
<Whatelay> 1610. The second element is certainly OE <léah>
'clearing (in a wood)'. The first may be <hwć:t> 'wheat' or
<wć:t> 'wet' (colon indicating a long vowel); on general
principles <hwć:t> is more likely, but there's really no way
to tell for sure. (The <woet> mentioned in the Wikipedia
article on Whatley doesn't exist.)

> However, I have had little success in confirming similar
> use in Normandy.

Because it's an English place-name; this family name
definitely does not go back to Normandy.

> This baffles me somewhat because our family history says
> that prior to Normandy, we came from Norway where we
> reportedly still have distant relatives using the name
> "hveitile" or "hvetileah".

To put it bluntly, this has to be family mythology, for
starters because family names in Norway aren't nearly old
enough. (They're significantly older in England.)

> For what its worth, our oral history says the name was
> initially given as a Norse descriptor for long, uneven
> blonde hair that looked like wild wheat.

That would be something like *hveitiligr. This would be a
very uncharacteristic formation: no Old Norse bynames in
<hveit-> or <-ligr> are known. The bynames <hvítr> 'white'
and <(hinn) hvíti> '(the) white', on the other hand, have
this meaning and are bog standard.

Brian

Brian M. Scott

unread,
Jan 12, 2008, 8:13:20 PM1/12/08
to
On Sun, 13 Jan 2008 01:23:17 +0100, Trond Engen
<tron...@engen.priv.no> wrote in
<news:9pednRcwxvV...@telenor.com> in sci.lang:

> WJW skreiv:

[...]

>> My surname is "Whatley", hence the reason for my search.
>> We know from family documents and oral history that the
>> family came to Wessex from Normandy during the Norman
>> Conquest. The name appears in Somerset as "Watelei"
>> and as "Watelege".

> The name looks unmistakenly English and a quick web search tells me that
> Whatley is the name of a village in Somerset
> (<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are Norman
> they certainly didn't bring that name with them.

> The name is supposed to mean "wet-glade".

Victor Watts, _The Cambridge Dictionary of English
Place-Names, prefers 'glade where wheat grows' and doesn't
even mention <wć:t> 'wet' as a possibility. I don't know
the book by Robinson cited in the Wikip. article, but Google
and the publisher's web site tell me that the long article
'English Language' (by Kenneth Turner, Michael MacMahon,
Nicola Woods, Olga Fischer, Wim van der Wurff, Richard
Coates, and Jacques Weber) in the Oxford journal 'The Year's
Work in English Studies' includes the following:

By contrast, Stephen Robinson's _Somerset Place
Names_ is best passed over in silence, but it is hard
to suppress a howl of outrage at a book shot through
with ...

I'm not about to pay for an 81-page article to find out
exactly what they had to say, but the gist is clear enough,
and I recognize several of the authors, especially Coates, a
place-name specialist, and Fischer.

[...]

Brian

WJW

unread,
Jan 12, 2008, 10:47:00 PM1/12/08
to
Brian & Trond,

Your assessment somewhat confirms my suspicion regarding the surname. Though
I know the family came from Normandy to Somerset, I would be willing to bet
that the surname was likely adopted in Somerset as a result of "de Watelei"
in terms of poll tax or otherwise. This would explain why I've been unable
to trace it in France. As for the oral history pertaining to blonde hair
(which is prevalent in our family to this day), well, I guess grandfathers
treasure the conveyance of such myths when little kids are sitting in their
laps on cold and snowy days.

Thanks again for all the excellent the feedback!

William

"Brian M. Scott" <b.s...@csuohio.edu> wrote in message
news:1tze9z4fcdv5g$.9zrywidmcdlc$.dlg@40tude.net...

Peter T. Daniels

unread,
Jan 12, 2008, 11:01:51 PM1/12/08
to
On Jan 12, 8:13 pm, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
> On Sun, 13 Jan 2008 01:23:17 +0100, Trond Engen
> <trond...@engen.priv.no> wrote in

> <news:9pednRcwxvV...@telenor.com> in sci.lang:
>
> > WJW skreiv:
>
> [...]
>
> >> My surname is "Whatley", hence the reason for my search.
> >> We know  from family documents and oral history that the
> >> family came to Wessex  from Normandy during the Norman
> >> Conquest.  The name appears in  Somerset as "Watelei"
> >> and as "Watelege".
> > The name looks unmistakenly English and a quick web search tells me that
> > Whatley is the name of a village in Somerset
> > (<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are Norman
> > they certainly didn't bring that name with them.
> > The name is supposed to mean "wet-glade".
>
> Victor Watts, _The Cambridge Dictionary of English
> Place-Names, prefers 'glade where wheat grows' and doesn't
> even mention <wæ:t> 'wet' as a possibility.  I don't know

> the book by Robinson cited in the Wikip. article, but Google
> and the publisher's web site tell me that the long article
> 'English Language' (by Kenneth Turner, Michael MacMahon,
> Nicola Woods, Olga Fischer, Wim van der Wurff, Richard
> Coates, and Jacques Weber) in the Oxford journal 'The Year's
> Work in English Studies' includes the following:
>
>    By contrast, Stephen Robinson's _Somerset Place
>    Names_  is best passed over in silence, but it is hard
>    to suppress a howl of outrage at a book shot through
>    with ...
>
> I'm not about to pay for an 81-page article to find out
> exactly what they had to say, but the gist is clear enough,
> and I recognize several of the authors, especially Coates, a
> place-name specialist, and Fischer.

Mike MacMahon is the Edinburgh phonetician, contributor to WWS, and
secretary of the IPA. Some sort of connection with April MacMahon, the
historical linguist, seems likely.

Brian M. Scott

unread,
Jan 13, 2008, 12:22:40 AM1/13/08
to
On Sat, 12 Jan 2008 20:01:51 -0800 (PST), "Peter T. Daniels"
<gram...@verizon.net> wrote in
<news:2e2d3ecb-5856-40f1...@j78g2000hsd.googlegroups.com>
in sci.lang:

[...]

> Mike MacMahon is the Edinburgh phonetician,

Glasgow.

> contributor to WWS, and secretary of the IPA. Some sort of
> connection with April MacMahon, the historical linguist,
> seems likely.

I thought of that, but she's McMahon, not MacMahon, and she
is at Edinburgh.

Brian

Dušan Vukotić

unread,
Jan 13, 2008, 5:22:22 AM1/13/08
to
On Jan 13, 1:23 am, Trond Engen <trond...@engen.priv.no> wrote:

> The name looks unmistakenly English and a quick web search tells me that
> Whatley is the name of a village in Somerset
> (<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are Norman
> they certainly didn't bring that name with them.
> The name is supposed to mean "wet-glade".

Similar to Serbian surname Vodelinić (wet-valley; wet-dell), Vodlić;
Vodli village in Moldavia (of Slavic origin), Russian personal Виталий
Vitaliy, supposed to be a form of Latin Vitalis (Ita. Vitale); Russ.
ветла vetla (willow, withy);

It is interesting to mention that Latin victus, vita -lis, vivus are
the words closely related to Serbo-Slavic ži-vljenje (living) and
Greek βιολόγος as well as it is related to English living. To
understand how it happened we must use Xur-Bel-Gon speech formula
(HSF) and start from the primal agglutinated form 'gibelgne' (from the
Gon-Bel-Gon ur-basis), wherefrom Serbian kobeljanje (roll about),
gibeljanje => gibanje (motion); hence gibeljanje => ži-bljenje =>
življenje (life);
From ži-bljenje sprang the Serbian word biće (being) and bihenje =>
biuenje => bivanje (existence; hence bivak bivouac).

Now we can also understand the history of the Latin vivo (from biva,
Greek βίος) and victus (from behagne, being, biće).

On the other side, English white is more complicated. If the word
'white' were related to Slavic svetlo (light) and svitanje (dawning)
then we would again go to the above mentioned Gon-Bel-Gon basis.
Namely, Slavic *svit- comes from *hosvit- (Serb. osvetljenje
illumination, liminance, lighting) or more precize from *gnobel- =>
gabel => *zabel- ; Serb. zabeleti whiten. Not incidantally, this
zabeleti (whiten) is related to the Serbian word nebo (sky); Serb.
"zora beli" (the dawn whitens). This zabeleti "shrank" later to zbelet
=> svelet => svelt => svet (world). We can see here that german Welt
could be a corresponding word to Slavic svelt => svet (world).

It is supposed that English "world" came from were-old (old man, "age
of man") but there is little credible evidence that it would come that
way. Some Germanic form similar to *huerelh' might have been existing
in earlear times. If it were true it would mean that the word "world"
was derived from Hor-Bel-Gon basis; i.e. from hurwelt => urwelt =>
wurwelt => wereld => world. It would at the same time explain the
difference between English world and german welt.

Even the Salvic word 'svet' (world) could be an 'descendant' of the
same Hor-Bel-Gon basis; i.e. it could be derived from the above-
mentioned 'zora bela' (white dawn); hurbelt => survelt => svelt =>
svet. Serbian word vrlet (craggy area); cf. Serbian syntagm "vrletan
svet" (worthy world);

At this momment we could return back to the beginning of this "story"
and "revise" the meaning of a supposed Englihs *were- (man) and Latin
virilis, which appears to be equal to Serbian vrlina (virtue). It
seems that English "worldly" and "worth(l)y" are the words derived
from the same place of origin?; similar as Serbian words 'vrletan'/
'vrli' (virile) and 'vredan' (worthy).

DV

Dušan Vukotić

unread,
Jan 13, 2008, 12:17:25 PM1/13/08
to
On Jan 12, 7:09 pm, Trond Engen <trond...@engen.priv.no> wrote:

> B&L quote Lithuanian <kvietỹs> as a Germanic loan. Obviously so, as
> Lithuanian would have šv- in an inherited word. I think the initial kv-
> dates the loan as either very old (Pre-Germanic) or fairly recent from
> some dialect with *hW- > /kv-/. Nowadays that feature is limited to West
> Scandinavian (including(?) Shetland) and some subgroup of Finland
> Swedish, and, since I'm not aware of a wider distribution historically,
> I'll go for Pre-Germanic. I suppose that the long y supports the
> reconstruction of a (Pre-)Germanic *-ija.

There are other Lithuanian words beside kvietys (wheat), like kviesti
(invite, ask) and kvitas (receipt). The corresponding Serbian words to
those Lithuanian could be 'cvet' (flower), kazati (say) and hvat/ati
(catch). All the words mentioned above are derived from the Gon-Bel-
Gon basis. We shall also see that words kviesti and kazati are
comparable to English ask (Serb. iskati, iskazati, Russ. skazat). Last
time I compared wheat with wet (Latin aqua, Serbian kvasiti) and now
it seems we can do the same with the Serbian verb okusiti (taste; from
okvasiti /soak, suffuse/ => okusti /taste/; cf. Latin gusto).
Nevertheless, this similarity appears to be accidental, thanks to the
common Gon-Bel-Gon basis those words were delivered from.

The common "denominator" for the "senses" in Serbian language is the
word čula (senses). Many unusual sound and morphological changes
occurred in the process in which a primitive agglutinated form close
to 'hneblhna' was "rearranged" into 'okolina' (surrounding; Lat.
accola neighbor and conlegium corporation, guild) and finally
"molded" into the modern Serbian word 'čula' (senses). Serbian
'slušati' (hear; from o/sluhnuti; Greek ακούω, ακουστικος acoustic)
sprang from the same source as 'čula'.

Sometimes, the meanings of certain senses do not share much with the
organs where that senses are placed. For instance, Serbian ukus
(taste; Lat. gustus) and verb okusiti (to taste) are more related to
oko (eye) and uho (ear) than to usta (mouth). At the same time
'usta' (mouth) are related to verbs okusiti (taste), 'ujesti' (bite)
and jesti (eat). The same case is with words kazati (say) and iskati
(ask). Serbian kazati (say) comes from ukazati (to point out) and it
means, "to perceive with eyes". Similar logic is in case of the word
okusiti (taste) because we need to say (kazati) what we have tasted.
In other words, our tasting organs are "telling" (kazati) us what are
we eating (jesti).

My above proposition that the Serbian words okusiti (taste) and kusati/
jesti (eat) might be related to aqua and Serbian verb
'ukvasiti' (soak) appeared to be completely wrong.

Let us now see what the relation could be between Lithuanian kvietys
(wheat) and Slavic cvet/kvet (flower). As we all know, there is know
visible wheat flowering and it suggests that these two words (kvietys,
kvet) cannot be directly related. On the other side, the process of
flowering is usually connected to white color (best visible on the
bloomed trees; English blossom; Lat. flos), what could lead us to
Slavic svitanje (dawning, whitening) and *svet-lo (light, visible,
white; svitanje (dawning) => cvetanje (flowering).


DV

Trond Engen

unread,
Jan 13, 2008, 1:16:26 PM1/13/08
to
Brian M. Scott skreiv:

> On Sun, 13 Jan 2008 01:23:17 +0100, Trond Engen
> <tron...@engen.priv.no> wrote in
> <news:9pednRcwxvV...@telenor.com> in sci.lang:
>
>> WJW skreiv:
>
> [...]
>
>>> My surname is "Whatley", hence the reason for my search.
>>> We know from family documents and oral history that the
>>> family came to Wessex from Normandy during the Norman
>>> Conquest. The name appears in Somerset as "Watelei"
>>> and as "Watelege".
>
>> The name looks unmistakenly English and a quick web search tells me
>> that Whatley is the name of a village in Somerset
>> (<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are
>> Norman they certainly didn't bring that name with them.
>
>> The name is supposed to mean "wet-glade".
>
> Victor Watts, _The Cambridge Dictionary of English
> Place-Names, prefers 'glade where wheat grows' and doesn't

> even mention <wæ:t> 'wet' as a possibility.

Because of the <wh->, I suppose. An important question, though: Do we
know anything about the length of the vowel? <hwæ:te> and <wæ:t> both
have long vowels and I can't read any length into the <a> of any of the
quoted forms of Whatley.

I'm still not all comfortable with the compound. Oakley, Ridgeley, and
even Oxley are OK, since their names may have stuck before settlement,
but there never were any <hwæ:te> until the <léah> was gone. Are there
any parallel examples?

I came to realize that my suggestion of last night, ON <hvatr>,
essentially means "sharp" (modern Norw. <hvass>/<kvass>) and is a
cognate of English <whet> v. "sharpen" (as in 'whetstone') < OE <hwæt>
a. "brave (< "sharp")". Thus the name could be a parallel to names like
Sharpley and Edgeley. That certainly could fit with the landscape,
either of the low ridge between the brooks where the village is situated
or, even better, of what became the quarry. But I don't know how to
account for the <-e->, so I guess <hwæ:te> is still the best alternative.

--
Trond Engen
- feeling a little whet, himself

WJW

unread,
Jan 13, 2008, 3:55:38 PM1/13/08
to
Trond,

> Because of the <wh->, I suppose. An important question, though: Do we know
> anything about the length of the vowel? <hwæ:te> and <wæ:t> both have long
> vowels and I can't read any length into the <a> of any of the quoted forms
> of Whatley.
>
> I'm still not all comfortable with the compound. Oakley, Ridgeley, and
> even Oxley are OK, since their names may have stuck before settlement, but
> there never were any <hwæ:te> until the <léah> was gone. Are there any
> parallel examples?

Right off the bat there is "Wheatley" and "Wheatleigh", both of which were
used by our family in Somerset and Wiltshire, and then later in Jamestown,
VA. In our family bible, there list Somerset surnames for family members as
Whateleia, Watele, Watileia, Wateuile, Watelega and Whatley. In fact,
Captain John Wheatleigh of Tingsboro, Somerset produced a will in 1530.
Within this document, he clearly referred to my son & heir as John
Wheatleigh, wherein the spelling remained the same. He then referred to my
brother Thomas Wheatly, thereby confirming variations in the spelling of the
surname, not only within the same family, but within the same document.
Later on, his grandson produces a will wherein he uses both the surnames
Wheatley and Whatley within the same document . . . while refering to
himself. There is also a written reference to a Elyas de Hwatele in1219
AssY, but this could just be a poll tax reference to his home (we do not
know if he was family). Likewise, there are written references to Henry de
Whateleia Hy 3 Colch (Essex), Walter de Whetele 1273 RH (Nottinghamshire),
Reginald de Watele 1296 and Robert Watteleghe 1296 SRSx. Again, we do not
know if any were family members and the "de" suggest they were not. What
we do know is that the surname Wheatley (and similar spellings) was/were
used by confirmed family members at various points in time.

>
> I came to realize that my suggestion of last night, ON <hvatr>,
> essentially means "sharp" (modern Norw. <hvass>/<kvass>) and is a cognate
> of English <whet> v. "sharpen" (as in 'whetstone') < OE <hwæt> a. "brave
> (< "sharp")". Thus the name could be a parallel to names like Sharpley and
> Edgeley. That certainly could fit with the landscape, either of the low
> ridge between the brooks where the village is situated or, even better, of
> what became the quarry. But I don't know how to account for the <-e->, so
> I guess <hwæ:te> is still the best alternative.

For what it is worth, John R. Clark Hall's "A Consise Anglo-Saxon
Dictionary" (1916) provides "hwet-stan" m. whetstone, AO. & "hwettan" to
'whet,' sharpen, incite, encourage, AE,CP. [hwaet]
(page 171).

William


WJW

unread,
Jan 13, 2008, 4:17:53 PM1/13/08
to
Trond . . . if I understand your thinking:

> I'm still not all comfortable with the compound. Oakley, Ridgeley, and
> even Oxley are OK, since their names may have stuck before settlement, but

> there never were any <hwć:te> until the <léah> was gone. Are there any
> parallel examples?

Going back to my initial questions, could this possibly be indicative of
"wild wheat growing in a woodland clearing or glade"?

Though people of the time most likely spelled to the best of their ability,
that ability was undoubtedly shaped by heritage and geographical context as
well as education. As such, the variations in spelling do not surprise me at
all and would be expected. However, one can be intelligent yet still not
literate so this makes me wonder whether word-play was used in forming some
of the surnames. If wheat is growing wild in a forest clearing away from an
intentionally planted field (hw?tecroft or hw?teland), then would it not be
possible that they called the location hw?tele or something akin? (hw?tele
> wh?tele > whatley).

William


Brian M. Scott

unread,
Jan 13, 2008, 4:43:27 PM1/13/08
to
On Sun, 13 Jan 2008 19:16:26 +0100, Trond Engen
<tron...@engen.priv.no> wrote in
<news:I5adnfKfPsb...@telenor.com> in sci.lang:

> Brian M. Scott skreiv:

>> On Sun, 13 Jan 2008 01:23:17 +0100, Trond Engen
>> <tron...@engen.priv.no> wrote in
>> <news:9pednRcwxvV...@telenor.com> in sci.lang:

>>> WJW skreiv:

>> [...]

>>>> My surname is "Whatley", hence the reason for my search.
>>>> We know from family documents and oral history that the
>>>> family came to Wessex from Normandy during the Norman
>>>> Conquest. The name appears in Somerset as "Watelei"
>>>> and as "Watelege".

>>> The name looks unmistakenly English and a quick web search tells me
>>> that Whatley is the name of a village in Somerset
>>> (<http://en.wikipedia.org/wiki/Whatley>). If your ancestors are
>>> Norman they certainly didn't bring that name with them.

>>> The name is supposed to mean "wet-glade".

>> Victor Watts, _The Cambridge Dictionary of English
>> Place-Names, prefers 'glade where wheat grows' and doesn't
>> even mention <wæ:t> 'wet' as a possibility.

> Because of the <wh->, I suppose. An important question,
> though: Do we know anything about the length of the
> vowel? <hwæ:te> and <wæ:t> both have long vowels and I
> can't read any length into the <a> of any of the quoted
> forms of Whatley.

<Whatelay> 1610 is late enough that it *may* imply original
length.

> I'm still not all comfortable with the compound. Oakley,
> Ridgeley, and even Oxley are OK, since their names may
> have stuck before settlement, but there never were any
> <hwæ:te> until the <léah> was gone.

Why not?

> Are there any parallel examples?

You mean compounds with words for cultivated crops? Beanley
(Nb), Barlow (Du), Flaxley (Gl, YW), Hailey (Bk), Lindley
(YW), Linley (Sa), Oteley (Sa), Ryley (La), from <béan>
'bean, pea, legume', <bere> 'barley', <fleax> 'flax', <hég>
'hay, mowing grass', <lín> 'flax', <áte> 'oats', <rýge>
'rye'.

[...]

Brian

Brian M. Scott

unread,
Jan 13, 2008, 4:59:09 PM1/13/08
to
On Sun, 13 Jan 2008 13:55:38 -0700, WJW <qua...@comcast.net>
wrote in <news:NPidnZiixZFW5xfa...@comcast.com>
in sci.lang:

[...]

> Likewise, there are written references to Henry de
> Whateleia Hy 3 Colch (Essex), Walter de Whetele 1273 RH
> (Nottinghamshire), Reginald de Watele 1296 and Robert
> Watteleghe 1296 SRSx. Again, we do not know if any were
> family members and the "de" suggest they were not.

It tells you absolutely nothing one way or the other. At
that date most toponymic bynames did still appear with the
documentary preposition <de>; by ~1400 they rarely did,
whether they were inherited or not. However, it's likely
that Henry's byname refers to Wheatley in Essex and Walter's
to Wheatley in Notts. Reaney clearly thought that
Reginald's and Robert's might refer to Wattlehill Farm in
Ewhurst, Sussex. If your family was clearly situated in
Somerset, these people probably didn't belong to it.

[...]

Brian

Trond Engen

unread,
Jan 13, 2008, 7:04:36 PM1/13/08
to
Brian M. Scott skreiv:

> On Sun, 13 Jan 2008 19:16:26 +0100, Trond Engen
> <tron...@engen.priv.no> wrote in
> <news:I5adnfKfPsb...@telenor.com> in sci.lang:
>

>> I'm still not all comfortable with the compound. Oakley,
>> Ridgeley, and even Oxley are OK, since their names may
>> have stuck before settlement, but there never were any
>> <hwæ:te> until the <léah> was gone.
>
> Why not?

Why uncomfortable? Because there's seemingly no time when the compound
might arise as a simple descriptive toponym. Why no wheat? Because it's
a domestic plant brought there and kept alive by farmers, and I couldn't
find any evidence of wild wheat in England. But it doesn't really matter
when ...

>> Are there any parallel examples?
>
> You mean compounds with words for cultivated crops? Beanley
> (Nb), Barlow (Du), Flaxley (Gl, YW), Hailey (Bk), Lindley
> (YW), Linley (Sa), Oteley (Sa), Ryley (La), from <béan>
> 'bean, pea, legume', <bere> 'barley', <fleax> 'flax', <hég>
> 'hay, mowing grass', <lín> 'flax', <áte> 'oats', <rýge>
> 'rye'.

... pretty much settles it. I gladly give in to evidence.

--
Trond Engen
- of Bowing Lowley

WJW

unread,
Jan 13, 2008, 7:57:09 PM1/13/08
to

> It tells you absolutely nothing one way or the other. At
> that date most toponymic bynames did still appear with the
> documentary preposition <de>; by ~1400 they rarely did,
> whether they were inherited or not. However, it's likely
> that Henry's byname refers to Wheatley in Essex and Walter's
> to Wheatley in Notts. Reaney clearly thought that
> Reginald's and Robert's might refer to Wattlehill Farm in
> Ewhurst, Sussex. If your family was clearly situated in
> Somerset, these people probably didn't belong to it.

This discussion has me taught me that the surname Whatley is most likely a
locative byname as you call it. The point I was focusing in on is that
"wheat" does occur again and again in the spelling variations, irregardless
of whether it is Somerset, Essex, Wiltshire, Gloucestershire, Staffordshire,
Warwickshire or Nottinghamshire. All of these locations also exhibit
spellings similar to Watelei and Whatley. Similar mixtures of these
spellings occur within single documents; some of which show the same person
using different spellings for his own name. Collectively, this provides a
good argument that the intended meaning behind the locative byname (and
subsequently the surname) is most likely wheat (hw?te) and not water, wet,
bold, sharp, quick or glorious. On a humorous note, I guess the surname
could possibly fit in your "grab-bag" of bynames meaning "what glade?"
(hwæt-leah? > what-ley > Whatley)

William

Andreas Prilop

unread,
Jan 14, 2008, 10:41:48 AM1/14/08
to
On Fri, 11 Jan 2008, WJW wrote:

> X-Newsreader: Microsoft Outlook Express 6.00.2900.3138


>
> Please excuse the "?" in my above post as I just realized the post did not
> carry the proper code.

I take your Subject to mean "Outlook Express assistance". ;-)

You need to choose

Tools > Options > Send
Mail Sending Format > Plain Text Settings > Message format MIME
News Sending Format > Plain Text Settings > Message format MIME
Encode text using: None

in order to send special, non-ASCII characters.
There is an explanation in German at
http://piology.org/news/oe-erste-schritte.html
I don't know of an English equivalent.

--
Bugs in Internet Explorer 7
http://www.unics.uni-hannover.de/nhtcapri/ie7-bugs

Brian M. Scott

unread,
Jan 14, 2008, 3:33:06 PM1/14/08
to
On Mon, 14 Jan 2008 01:04:36 +0100, Trond Engen
<tron...@engen.priv.no> wrote in
<news:Pr-dncfk7di...@telenor.com> in sci.lang:

> Brian M. Scott skreiv:

>> On Sun, 13 Jan 2008 19:16:26 +0100, Trond Engen
>> <tron...@engen.priv.no> wrote in
>> <news:I5adnfKfPsb...@telenor.com> in sci.lang:

>>> I'm still not all comfortable with the compound. Oakley,
>>> Ridgeley, and even Oxley are OK, since their names may
>>> have stuck before settlement, but there never were any

>>> <hwć:te> until the <léah> was gone.

>> Why not?

> Why uncomfortable? Because there's seemingly no time when
> the compound might arise as a simple descriptive
> toponym. Why no wheat? Because it's a domestic plant
> brought there and kept alive by farmers, and I couldn't
> find any evidence of wild wheat in England.

I understood that; I just didn't see why it should be a
problem, since the use of the glade for growing wheat might
reasonably have been the reason for giving it a name in the
first place.

[...]

Brian

Trond Engen

unread,
Jan 14, 2008, 5:04:50 PM1/14/08
to
Brian M. Scott skreiv:

> On Mon, 14 Jan 2008 01:04:36 +0100, Trond Engen
> <tron...@engen.priv.no> wrote in
> <news:Pr-dncfk7di...@telenor.com> in sci.lang:
>
>> Brian M. Scott skreiv:
>
>>> On Sun, 13 Jan 2008 19:16:26 +0100, Trond Engen
>>> <tron...@engen.priv.no> wrote in
>>> <news:I5adnfKfPsb...@telenor.com> in sci.lang:
>
>>>> I'm still not all comfortable with the compound. Oakley,
>>>> Ridgeley, and even Oxley are OK, since their names may
>>>> have stuck before settlement, but there never were any

>>>> <hwæ:te> until the <léah> was gone.

>
>>> Why not?
>
>> Why uncomfortable? Because there's seemingly no time when
>> the compound might arise as a simple descriptive
>> toponym. Why no wheat? Because it's a domestic plant
>> brought there and kept alive by farmers, and I couldn't
>> find any evidence of wild wheat in England.
>
> I understood that; I just didn't see why it should be a
> problem, since the use of the glade for growing wheat might
> reasonably have been the reason for giving it a name in the
> first place.

Thanks. I prefer that kind of straightforward naming, but I got lost in
the forest trying to find a literal sense of "glade". My OERD explains
modern <lea> as "a piece of meadow or pasture or arable land". The
obvious point that totally slipped me is that the use of the glade for
growing wheat might have come before people set up their homes there.
The simple semantic drift from "natural opening in the forest" to "used
land" is the linguistic evidence of this period of expansion into the
forest.

--
Trond Engen
- simple semanticist

0 new messages