(Preliminary note : I use the sign # to mark elements attested only

inflected or compounded.)



_Sharbhund_ appears in the _Narn i [Ch]în Húrin_ (UT:98) as the

Dwarvish name of _Amon Rûdh_ in use among the Petty-Dwarves.

There is no  explanation for this word; however, it is a frequent

practice in Middle-earth to give a place different names according

to the language, but with the same meaning. Therefore it is not unwise

to think that _Sharbhund_ means the same as _Amon Rûdh_, i.e.

'Bald Hill' (cf. UT and Silmarillion Indexes).



The problem is how to segment this word. Till now I have mostly

seen it divided as _Shar-bhund_,  for instance by Lisa Star in her

"Khûzdul or Dwarvish wordlist"

(http://www.geocities.com/Athens/Parthenon/9902/khuzdul.html),

Helge Fauskanger on his Ardalambion article "Khuzdul - the secret

tongue of the Dwarves"

(http://www.uib.no/People/hnohf/khuzdul.htm),

or Edouard Kloczko in his "Dictionnaire des langues des Hobbits,

des Nains, des Orques" (A.R.D.A., Argenteuil, France, 2002 -

ISBN  2-911979-04-4). All refer to a possible link bewteen the

_#bhund_ element and  the word _bund_ "head" (VII:174). Edouard

Kloczko propounds the hypothesis of an  augmentative infix _h_ as

he compares our form with other Dwarvish  words. Nevertheless I

think that a division as _Sharb-hund_ is  more likely for two

reasons.



The  first is phonological: _#bhund_, if really derived from _bund_

by an _h_ infix,  would display an initial consonant cluster, a thing

which is not found in our  Khuzdul corpus (th and kh seem to behave

as unitary phonemes). True, this  argument is weakened by the limited

data, so we cannot be entirely sure that  this restriction is valid; all the

same, I do not like the idea of positing  morphemes with an unattested

structure. Incidentally, we know that Adûnaic,  which is said to have

been influenced at an early stage by Khuzdul (named  "Khazadian" in

the source, i.e. Lowdham's Report), had no initial consonant clusters

either. Neither does Quenya, though here the definition of a consonant

cluster is admittedly problematic; cf.  IX:418.



The  other is structural: there is a general agreement that Dwarvish

roots are mostly  triconsonantal, with a smaller number of biconsonantal

roots; Lowdham says in  IX:415 that in its predominance of triconsonantal

roots "Adûnaic shows affinity  with Khazadian". Therefore a division in

_Sharb-hund_ appears morphologically  simpler and more probable: it

would imply two classic triconsonantal roots  *Sh-R-B and *H-N-D. The

second element has a structure 1u23 (where 1-2-3 are the  radicals) attested

in _bund_ "head" (VII:174) and _Rukhs_ "Orc" (XI:391). We  have no isolated

word identical in structure with the first element, but we have  something

quite similar in _Mazarbul_ if we take off the _-ul_ ending, which is quite

clearly a suffix used to denote a relationship (cf. _Fundinul_ "[son] of

Fundin" of the Moria gravestone, _Khuzdul_ "Dwarvish").



Since  the composition order seems to be determiner-determined in

Dwarvish (cf. the  list of examples in Ardalambion's article mentioned

above), we could deduce the  words _#sharb_ "bald" and _#hund_ "hill",

provided that _Sharbhund_ is indeed  semantically equivalent to_Amon

Rûdh_.



Bertrand  Bellet



-----------------------------------------------------------

Language  has both strengthened imagination and been freed by it.

Who shall say whether  the free adjective has created images bizarre

and beautiful, or the adjective  been freed by strange and beautiful

pictures in the mind ? - J.R.R. Tolkien,  A Secret  Vice







[Non-text portions of this message have been removed]

--- In lambengolmor@yahoogroups.com, BertrandBellet75@a... wrote,

regarding _Sharbhund_, Dwarvish name of _Amon Rûdh_ 'Bald Hill':



> (Preliminary note : I use the sign # to mark elements attested only

> inflected or compounded.) ... We  have no isolated word identical

> in structure with the first element [presumably _Sharb-_ according

> to Bertrand], but we have something quite similar in _Mazarbul_ if

> we take off the _-ul_ ending, which is quite clearly a suffix used to

> denote a relationship (cf. _Fundinul_ "[son] of Fundin" of the Moria

> gravestone, _Khuzdul_ "Dwarvish").



Strengthening this particular note of Bertrand's, I'd like to comment

that the deduced word *_mazarb_ probably functions according to

Bertrand's conception -- it seems that the radical word is #_zarb_

while the prefix _ma-_ may be instrumental (meaning the root Z-R-B

means "write, record" or something similar), and also being an "ism"

by the Professor -- who knew Hebrew to some extent. In Hebrew this

prefix is very eminent (along with its allomorphs, _ta-_ as in

*_tawrawat_ > _Torah_ and its passive allomorph _mu-_ as in

_mutar_ "allowed", root TWR), like in the word _mazon_ "food" from

the root ZWN "feed".



John

--- In elfling@yahoogroups.com (message #30489), David Giraudeau wrote:



> In VT46, appendix 3, page 30, I think Mr Smith would have made a mistake

> by putting the tengwa #16 (tengwa of "Unque") instead of the tengwa #8

> (tengwa of "Ungwe"), in the paragraph beginning by "Five other names...".



I post this message here since its content seems to be relevant to the goals of

the list.



Sébastien Bertho



{(Dr.) Arden Smith replies: "Quite right.  The tengwar in the article were

changed from one font to another after I submitted the appendix (my original

definitely had ungwe), but I should have caught the error at the proofreading

stage."]

My post is primarily concerned with ambiguities caused by hyphens occuring in

some Elvish words in the A&C (VT45 & VT46), and also in the published text of

the _Etymologies_ (V:347-400). There are some cases where hyphens appearing at

the end of lines might be mere line's end rules or real cases of caesuras,

denoting compound words. So I'd like to take a look at all the occurences of

sensitive hyphens throughout the A&C. (Note that I replace original macrons in

the text with circumflex accents for reasons of convenience.)



[We thank Sébastien for compiling this list of examples of hyphens and other

typographical features of the published _Etymologies_ as well as our "Addenda

and Corrigenda" to same that are potentially ambiguous. As Sébastien has

observed, the page-layout software used to produce _Vinyar Tengwar_ was

inadvertantly configured to apply its normal hyphenation rules to accommodate

line balancing, with the result that hyphens were placed in Elvish forms where

they do not occur in the MS. To address these instances and ambiguities, we

append our clarification of all points raised after each entry. Please note that

we use a circumflex above a vowel to indicate a macron in the MS and published

texts, due to the limitations of the Yahoo groups intereface. CFH/PHW]



- VT45:6 s.v. BAD[2]-, we have two occurrences of _Anga-vanda_. But since

_Angavanda_ is given under ANGÂ- on the same page, I suppose that it is the same

form under BAD[2]-, without hyphen. [The MS reading is _Angavanda_ in both

occurrences s.v. BAD[2]-. CFH/PWH]



- VT45:8 s.v. DA3-, we have the form _dae-deloth_ and between pp. 8 and 9, s.v.

DAY-, we have also _Dae-dhelos_. The published text gives _Daeðelos_ and

_Daedhelos_ (V:354-55, s.v. DAY- & DYEL-), and we have also _Daedhelos_ under

DYEL- in the A&C (VT45:11). Hence, I assume that the second form is almost

certainly _Daedhelos_, but I'm not sure for the first (under DA3-). Is it

_dae-deloth_ or _daedeloth_? [The MS reading is _dae-deloth_ s.v. DA3-, and

_Daedhelos_ s.v. DAY-. CFH/PHW]



- VT45:11 s.v. DUL-, we have _Ter-endul_ in the editorial note, whereas the

beginning of the entry states that the form _Terendul_ in the published text

(V:355)  should be _Teren-dul_. This is indeed a sensitive hyphen case! But we

also have _Terendul_ in the editorial note, two lines below _Ter-endul_, so I

think that in this case the hyphen is only a rule. [The MS reading is

_Terendul_, not _Ter-endul_. CFH/PHW]



- VT45:11, s.v. DYEL-, we have _Del-duthling_, but this is clearly a case of

loose compound since the correction bears on the hyphen ([for:] _Delduthling_

[read:]_Del-duthling_). [Correct; the MS reading is _Del-duthling_. CFH/PHW]



- VT45:13 s.v. GAL1- we have _glan-tha-_ in the editorial note, but this is

undoubtedly a case mere rule (_glantha-_). [The MS reading is _glantha-_.

CFH/PHW]



- VT45:20 s.v. KEPER-, we have _Sarn-gebir_ (>> _Serngebir_), but the published

text (V:363) and the editorial notes refer to the other form _Sern Gebir_. So,

is it _Sarngebir_ or _Sarn-gebir_? [The MS reading is _Sarngebir_. CFH/PHW]



- VT45:26 s.v. LÁYAK-, _Lhe-benedh_ is certainly _Lhebenedh_ (the N stem is

_lhoeb_, _lhaeb_ or *_lheb_ and there's no reason to suppose a breaking of the

stem at this point). [The MS reading is _Lhebenedh_. CFH/PHW]



- VT45:30 s.v. MA3-, _matthô-be_ (with an acute accent on the _ô_) is

undoubtedly a compound, like the reading it replaces (_matthô-be_). [The MS

reading is _matthô-be_ (with an acute accent over the _ô_. CFH/PHW]



- VT45:32 s.v. MAT-, _morn-magli_. In this case (an historic etymological form),

I suppose that we are confronted with a (loose) compound. [The MS reading is

_morn-magli_. CFH/PHW]



- VT45:35 s.v. MOR-, the form _muri-lind_ in the editorial note seems evidently

to be a repetition of _murilind_, given two lines above. [The MS reading is

_murilind_. CFH/PHW]



- VT45:37 s.v. NÁYAK, the form _naiqe-lea_ is undoubtedly a single word (like

_naiqele_ or the later form _naikelea_). [The MS reading is _naiqelea_. CFH/PHW]



- VT45:38 s.v. NDÛ-, we have _nû-men_ in lines 3 & 4 of the editorial note,

apparently a repetition of the compound _nû-men_ given two lines above. [The MS

reading is _nû-men_. CFH/PHW]



- VT46:4 s.v. ÑGOROTH-, _Nann Dun-gorthin_ is undoubtedly _Nann Dungorthin_ (the

correction bears on _Nann_ for published _Nan_). [The MS reading is _Nann

Dungorthin_. CFH/PHW]



- VT46:7 s.v. ORO-, _oro-drim_ is evidently _orodrim_, as in the published text.

[The correct reading is _orodrim_. CFH/PHW]



- VT46:10 s.v. RAB-, _rho-fannor_ is obviously _rhofannor_ (like _rhofan_). [The

MS reading is _rhofannor _. CFH/PHW]



- VT46:15 s.v. SPAN-, we have _span-turo_, and the published text gives

_Span-turo_ (V:387 s.v. SPAN-). What about the capital (and the hyphen)? [The MS

reading is _span-turo_. CFH/PHW]



- VT46:18 s.v. THIN-, _Tor-thingol_ is obviously a compound since it replaces

the reading _Torthingol_ in the published text. Under the same entry, _thin-tha_

is on the other hand obviously a case of rule hyphen, since the published text

gives _thintha_. [The MS readings are _Tor-thingol_ and _thintha_. CFH/PHW]



I'd like also to point some cases of sensible hyphens occuring in the published

text which are not alluded to in the A&C:



- V:353 s.v. BOR- : _voro-gandele_ is a case of rule since VT45:7 s.v. BOR-

states that **_vorogandele_ in the published text stands for _vorongandele_.

[The MS reading is _vorongandele_. CFH/PHW]



- V:355 s.v. DUN- : _Nan Don-goroth_ seems to be a mere case of rule (it is

listed together with _(Nan) Dungorthin_ and _Nann Orothvor_). [The MS reading is

_Nan Dongoroth_. CFH/PHW]



- V:359 s.v. GOND- : _Gon-dothrim_ is obviously a single word (like the related

form listed after, _Gondothrimbar_). [The MS reading is _Gondothrim_. CFH/PHW]



- V:360 s.v. 3AR- : _Gar-thurian_ seems to be a single word since it is based

upon Dor. _garth_ (and _Garthurian_ appears in the first, deleted, version of

this entry in VT45:16 s.v. 3AR-). [The MS reading is _Garthurian_. CFH/PHW]



- V:368 s.v. LAT- : _Tumbo-latsin_ may be a unitary word or a loose compound.

[The MS reading is _Timbolatsin_. CFH/PHW]



- V:369 s.v. LIN2- : _Laure-lingen_ seems to be a single word (a variant of

_Laurelinden_), but it could be a compound explaining the other interpretation

of the name _Laurelin_. [The MS reading is _Laurelingen_. CFH/PHW]



- V:372 s.v. MBAR- : _Em-merein_ seems evidently to be a single word (cf. the

form given just after, _Emerin_). [The MS reading is _Emmerein_. CFH/PHW]



- V:376 s.v. NEL : Dor. _gal-breth_ (obviously a single word since it is already

preceded by an etymological form _galdbreth_).  [The MS reading is _galbreth_.

CFH/PHW]



- V:380 s.v. PEL- : _*pel-takse_ is an historic etymological form and may be a

loose compound. [The MS reading is *_pel-takse_. CFH/PHW]



- V:381 s.v. PHIN- : *_Phinde-rauto_ is an historic etymological form and may

well be a compound. [The MS reading is _Phinde-rauto_. CFH/PHW]



- V:389 s.v. TÂ-, TA3- : _Tinwe-rontar_ is likely to be a case of rule hyphen

(the name appears as _Tinwerontar_ under TIN-). [The MS reading is

_Tinwerontar_. CFH/PHW]



- V:391 s.v. TEL-, TELU- : **_tel-manna_ is evidently a single word _talmanna_

(cf VT46:18: "[for:] _telmello telmanna_ [read:] _telmello talmanna_").

[Correct. CFH/PHW]



- V:393 s.v. THUR- : _garð-thurian_ is probably a loose compound (explaining the

etymology of _Garthurian_), and this is probably also the case of _Ar(ð)-thoren_

(etymological form of _Arthurien_). _Thurin-gwethil_ may be a single word (the

name is referred to as _Thuringwethil_ three lines below). [The MS readings are

_garð-thurian_, _Ar(ð)-thoren_, and _Thuringwethil_, respectively. CFH/PHW]



- V:393 s.v. TIN- : _Tin-wetar_ is obviously a case of line's end rule (this

name is made of _tinwe_ and _tar_), and it is very likely to be the same for

_Tin-dômiselde_ (the form _Tindômiselde_ appears under the stem SEL-D- and is

apparently made of _*tindômi_ and _*selde_). [The MS readings are _Tinwetar_ and

_Tindômiselde_ (ô representing o-macron). CFH/PHW]



Another thorny issue is concerned with _ae_ and _oe_, or more precisely with _æ_

and _œ_ (_ae_ and _oe_ joined together), when they appear in lower-case. There

are several forms, whether in the published text or in the A&C, implying one of

these characters. The problem is that these forms are all in italic script, and

with the fonts used it is sometimes difficult to tell which of them is intended

(most especially in the published text of the _Etymologies_). This problem is

indeed tricky since this is an issue in which Tolkien hesitated. I think it

should a good thing to clarify things once and for all in this matter to avoid

all subsequent confusions. (I give intentionally the digraph in brackets after

the single letter, for the latter may cause problem for some people, depending

on their language settings, operating system and/or character encoding, but it

must be understood that the words quoted only involve the single character.)

Entries are listed in alphabetical order, whether coming from the published text

or from the A&C.



- VT45:8 s.v. DAR-, we have apparently _æ_ (ae) in _dæri_ (daeri). [The MS

reading is (ae). CFH/PHW]



- VT45:11 s.v. DÓRON-, we seem to have _œ_ (oe) in _dœrœin_ (doeroein) and

_derœin_ (deroein). [The MS reading is (oe) in both cases. CFH/PHW]



- VT45:11 s.v. DUL, this is obviously _œ_ (oe) in _dœlio_ (doelio) since the

correction bears on _oe_/_œ_ ("[for:] _doelio_ [read:] _dœlio_"). [The MS

reading is (oe). CFH/PHW]



- VT45:13 s.v. GAL1-, apparently _œ_ (oe) in _gœlia-_ (goelia), though it might

perhaps be _æ_ (ae), since the root vowel is A. [The tentative reading given in

the A&C is (oe), though it is also possible to read the MS as (ae). CFH/PHW]



- V:358 s.v. GÁYAS-, it seems we have _æ_ (ae) in ON _gæsra_ (gaesra), with a

macron on the digraph (impossible to write in standard characters without

requiring to unicode or special fonts), since the stem is in -A- and the N

cognate is _gaer_. [The MS reading is (ae). CFH/PHW]



- V:362 s.v. KAL-, we seem to have _æ_ (ae) in the etymological form _*gâlæ_

(*gâlae). [The MS reading is uncertain, but most likely (ae). CFH/PHW]



- VT45:24 s.v. KWÆ-, apparently _æ_ (ae), in _kwæ-nê_ (kwae-nê), with a macron

on the digraph. [The MS reading is (ae). CFH/PHW]



- V:375 s.v. NDER-, we seem to have _æ_ (ae) in Eldarin _ndær_ (ndaer), with a

macron on the digraph, since the Q deivative is _nêr_ and the ON cognate is

_ndair_ (although the N form is _doer_). See also below (NÎ1-). [The MS reading

is (ae). CFH/PHW]



- V:378 & VT46:4 s.v. NÎ1-, _œ_ (oe) in _ndœr_ (ndoer), with a macron on the

digraph, replaced by _ndær_ (ndaer), _æ_ (ae) with a macron. [Correct. CFH/PHW]



- VT46:6 s.v. NOT-, we obviously have _œ_ (oe) in _nœdia_ (noedia) and

_arnœdiad_ (arnoediad). [The MS reading is (oe) in both cases. CFH/PHW]



- VT46:10 s.v. PHOR-, we evidently have _œ_ (oe) in _fœir_ (foeir). [The MS

reading is (oe). CFH/PHW]



- VT46:12 s.v. RUN-, we obviously have _œ_ (oe) in _tellœin_ (telloein). [The MS

reading is (oe). CFH/PHW]



- V:387 s.v. SNAS-, SNAT-, it seems we have _æ_ (ae) in Dan. _snæs_ (snaes),

with a circumflex on the character (cp. with the N cognate _naith_). [The MS

reading is (ae) with circumflex. CFH/PHW]



- VT46:18 s.v. TEÑ-, we apparently have _æ_ (ae), in _tær_ (taer), and in _tæma_

(taema), the latter with a macron on the digraph. [The MS reading is (ae) in

both cases, with a macron over the latter case. CFH/PHW]



- VT46:19 s.v. TIN-, we evidently have _œ_ (oe) in _mœrilinn_ (moerilinn) and

_mœrilind_ (moerilind). [The MS reading is (oe) in both cases. CFH/PHW]



- VT46:20 s.v. ULU-, we evidently have _œ_ (oe) in _œil_ (oeil). [The MS reading

is (oe). CFH/PHW]



- V:397 s.v. WED-, it seems we have _æ_ (ae) in _*wædê_ (*waedê), with a macron

on the character. [The MS reading is (ae) with a macron. CFH/PHW]



I've tried to be as exhaustive as possible on the topics of hyphens and ae/oe.

Feel free to point out to the list any occurence that I might have missed.



Some other issues concerned with the _Etymologies_ and the A&C :



Under the entry ERÉK- in the published Etymologies (V:356), the reference to the

stem OR-NÍ is a typo. Under the stem ORO- (V:379), the extended root having to

do with high trees is ÓR-NI (with the acute accent on the O) and there is also a

reference to ÓR-NI under TUS- (V:395, end of the entry). In VT46:7 (s.v. ORO) it

is also said that "the base ÓR-NI was first written as ÓR-ON". [Correct.

CFH/PHW]



In VT45:35 s.v. MIZD-, the Dor. adj. _méd_ given in the published text is said

to be read as _mêd_ (with a macron). But in the comments for this entry, it is

said that the gloss of _mêd_ (with a circumflex accent) was added later. Is the

form in the MS given with a macron or with a circumflex? [Dor. _mêd_ (with a

circumflex) in the comments is an already-corrected typo for _mîd_ (with

macron); see the VT errata at <http://www.elvish.org/errata/>. CFH/PHW]



In VT45:37 s.v. NAY-, we have "[for:] _nae_ alas [read:] _nae_, _nae_ alas". The

repetition of _nae_ without any  difference between the two forms seems weird.

Is it really the reading given in the MS ? Could it be possible that one of

these forms contain _ae_ joined together and not the other, or that one contain

_ae_ and the other _oe_? [The reading with repetition is correct; it apparently

reflects the repetitious form of the phrase, not uncommon in lament, as e.g.

English "no, no!" CFH/PHW]



In VT45:37 s.v. NDI- (DI-), we have the sentence _tiro men. di ngorgoros_ (or

_ngorgoroth_), evidently a sketch preparatory to the song "_A Elbereth_" in LR.

Is the reading of the dot certain ? Could it rather be a median dot (·), as seen

in many sentences (i.e. *_tiro men·di ngorgoros_)? [The dot is in fact a period,

as per the A&C, the two parts of the phrase being written in this draft form in

two steps, the first originally ending a complete thought. CFH/PHW]



In VT45:28 s.v. LOP-, OLOP- 'horse', it is said that the entry was originally

written as LOB-. It is then noted in the comment that the Q form _olombo_, left

unchanged when the stem was changed from LOB- to LOP-, OLOP-, could not derive

from the latter. Why not? Of course, if we imagine a strengthening of the last

consonant of the stem (or a metathesis after the addition of a suffix _-mo_ or

_-no_), the most probable solution seems to be *_olompo_, since the regular

pattern observed for stems ending in -P is, in these cases, P > MP (we have no

example of P > MB in the _Etymologies_). Hence the form _olombo_ could not

derive straight from LOP-, OLOP-. But the change _*olopo_ > _olombo_ or

_*olop-mo_ (or _*olop-no_) > _olombo_ seems not unlikely and might have occurred

by analogy (with some stems in -B or -M which produce strengthened forms in -MB)

or by dissimilation. [As Sébastien notes, there are no examples of P > MB in the

_Etymologies_, and this fact adequately justifies our statement that "_olombo_

... could not derive from the base LOP-." Certainly the proposal that _olombo_

might derive from *_olop-no_ is demonstrably false, since in the _Etymologies_

original -PN- yields -MN- in Qenya; note Q. _telemna_ 'silver' (V:367), prob. <

*_telep-na_; and Q. _lemnar_ 'week' (of five days) (V:368), prob. < *_lep'n-ar_

< *_lepen-ar_ 'five days' (cp. Q. _are_ 'day').]





Sébastien Bertho