PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Send a file File manager PDF Toolbox Search Help Contact



albanian PIE to friedrich .pdf



Original filename: albanian PIE to friedrich.pdf

This PDF 1.4 document has been generated by / iText® 5.5.5 ©2000-2014 iText Group NV (AGPL-version); modified using iText® 5.5.5 ©2000-2014 iText Group NV (AGPL-version), and has been sent on pdf-archive.com on 13/12/2015 at 10:47, from IP address 104.156.x.x. The current document download page has been viewed 276 times.
File size: 651 KB (8 pages).
Privacy: public file




Download original PDF file









Document preview


Albanian corrigenda to Friedrich's Proto-Indo-European trees
Author(s): Martin E. Huld
Source: Zeitschrift für vergleichende Sprachforschung, 95. Bd., 2. H. (1981), pp. 302-308
Published by: Vandenhoeck & Ruprecht (GmbH & Co. KG)
Stable URL: http://www.jstor.org/stable/40849473
Accessed: 13-12-2015 09:39 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

Vandenhoeck & Ruprecht (GmbH & Co. KG) is collaborating with JSTOR to digitize, preserve and extend access to Zeitschrift
für vergleichende Sprachforschung.

http://www.jstor.org

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

Albaniancorrigenda
trees
to FriedricheProto-Indo-European
By examiningthe arborealsystemof the Indo-EuropeansPaul
Friedrich(Proto-Indo-European
trees,1970) has made a significant
contribution
to the study of Indo-Europeanlinguisticdomains.
In the course of his work Friedrichdrew most of his Albanian
der albanesischen
materialfromMeyer'sEtymologisches
Wörterbuch
and
Jokl's
masterwork,
Linguistisch-kulturhistorische
Sprache(1891)
aus demBereichedesAlbanischen
(1923). For these
Untersuchungen
bothJokland Meyeremployeda philologicaltranscrippublications
tiondevisedin thenineteenth
century.It was notuntilafterWorld
- officiallyadopted in
War II that the standard orthography
1912 - came into widespreaduse. As a result of this change,
muchofthat materialFriedrichcitesis in an outmodedtranscription whichdue to its peculiarnatureis proneto misspellings
by
These same problemsconfronted
the non-specialist.
Pokornywhen
etymologisches
citingAlbanian materialfor the Indogermanisches
for the latterwork Hamp publishedan eight-page
Wörterbuch;
list of corrections
(IF, 67 [1962]: 142-50).
With the view of similarlyimprovingthe usefulnessof FriedI proposeto add a few corrich'smaterialforthe non-specialist,
rectivenotesto the thirty-item
synoptictable (pp. 24-25) and to
the list of miscellanicaarbórea(149-51). For simplicityI will cite
then the Albanian formto be disFriedrich'sIE reconstruction
cussed as it appears in the spellingof the Fjalor i gjuhësshqipe
(Dictionaryof the Albanian language,Tiranë, 1954) along with
fromthe Fjalor. For
an Englishgloss and the botanicalreference
Friedrich'scitationis givenin brackets.These are
cross-reference
followedby a fewcomments.It mustbe stressedthat these commentsare oftententativefortherehas been no complete,welldocumentedsurveyof Albanianplant-namesto supersedeJokl's
discussion(1923: 177-235).
In citingAlbaniannouns,feminines
end in an unstressedcentral
is
vowel written-ë, which replacedby -a in the nominativesingular definite.In some dialects this phoneme has become lost
resultingin apparent identitywith indefinitemasculinenouns;
masculinenouns are identifiedby the suffixeddefinitearticle-i

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

Albanian corrigendato Friedrich's Proto-Indo-Europeantrees

303

(-u after-g, -k, -A,or a stressedvowel). Adjectivesare indicated
by a preposedconcordparticle,masculinei femininee.
*bherHgo-:No cognateis givenin the table,but Friedrichgives
bothbredhand bardhëin thetext,i, e bardhë'white'is a zero-grade
from*bh?H§o-,a formalso
exactlymatchingOIn bhürja-cbirch9
attestedin Lat frãxinusfash' α-coloringis the resultof a long
The
(laryngealized)syllabicresonant,cf. pare 'first'< *ρχχ-ηο-.
is
also
indicated
Lithuanian
the
intonation,
laryngeal
by
full-grade
-i cfir,Abies', as Friedrichnotes,
bérïas'birch' < *bherH$o-.
bredh,
is a notoriousproblem.If it is relatedto *bherH§o-,
bredhmay be
an earlymetathesisforthe expected*bierdh.On the otherhand
bredhand Rum brad'Scots pine' whichis probablya loan from
Albanianmay,like the Latin hapax bratus(Pliny) and Gk. βράϋν
'savin', reflectloans fromthe Near East best equated with Heb
Arambrotboth'cypress'and Akkad G^burãéu'juniper' <
b(d)rõé9
Proto-Semitic*brãdu.A relationshipwith OE brord'needle of a
plant', ON broddr'(pine)needle'< PG *bruzdaz< PIE *bhf8dho(Jokl,IF 30 [1912]: 209) must also be considered.
*pytw-:pishë '(Scots) pine, Pinus' [piëè, ? an errorfor piêë,
a partial modernizationof piSe], The protoformis *piitso-and
sh
matchesthe Latin substantizedadjective plnus < *piits-no-,
frommedial*-ts-is regular,cf.mot'season,year' : : moshë'old age'.
cf. Gk.
There also occurspisë 'bitumen,hell' reflecting
*pilc-ieh,
πίσσα'tar'.
The question
*maHlo-:molle'apple,Pirus (sic) malus'[moZZe(?)].
markdenotesa well-founded
doubtwhetherthisis a nativeterm
or a loan fromLat mãlumor NorthwestGreek*μαλον.In some
contextsmolla(pi) denotes'cheeks'. The agreementbetweenthis
use and TA malan(notmalan).TB meli(pi) 'nose',Gk.μήλα'cheeks,
breasts',Lat mala 'cheekbone',perhaps a reinterpretedneuter
plural,is too generalto be of consequence;only Hitt mahla-'?
grapevine'failsto show this semanticassociation.
*alyso-: Friedrichdoes not list an Albanian cognate for the
alder terms.There are two subsets in Indo-European;the first
comprisesGermanic*eliza,Slavic *olixa9Celticplace-names,and
the Makedoniangloss αλιζα-ή
(the whitepoplar
λεύκητωνδένδρων
a
These
forms
to
PIE
*aliso-,
Friedrich(70)
point
amongtrees)1).
2) The qualifyingphrase των δένδρωνsurely indicates that λενκηis to
be taken as 'poplar* and not 'leprosy*as Katiöic (Ancientlanguages of the
Balkans, 1976) does.
20 Zeitschriftfür vgl. Sprachforschung,
Bd. 95, Heft 2

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

304

MartinE. Huld

noted Szemerényi'squalms over derivingLat alnus 'alder' from
as a 'morphological
characterized
*alsno-,a proto-form
Szemerényi
monster*.
These objectionsare speciousfortwo reasons.The proposed alternative*alisno-would give Lat *alïnus ratherthan
alnus, cf. ahênus'bronze9< *ayesnos;and not only does Indoconsonantclustersbut
Europeanhave a penchantfor'monstrous9
Lith alksnis(also elksnis,cfLatv èlksnis)showsjust sucha cluster
in this very stem, alnus could representa thematic*al(Jc)snobeside the Baltic i-stems(froma consonantstem?). The initial
α-vocalismsuggestsa laryngeal.If this were *A29the proto-form
accountsnot onlyforLat alnus but Alb halë 'black or
*hel(k)snoAustrianpine, Pinus nigra92).Conifersformone of the largest
ofAlbanianforests,
withbeechesforsecond
components
competing
place afteroaks. Pine predominatesin the north,spruceand fir
in the south.The identification
of the 'alder9as a pine or spruce
have
the
been
cone-likefruitsof the formeras
may
suggestedby
well as the comparativerarenessof the alder in the Balkans. In
Albanianthe alder is designatedas in Celticby the root *urn-z).
Homophonouswithhalë 'black pine9is hale 'splinter9denoting
a largenumberof splinteryitemssuch as wood chips,fish-bones,
awns etc. At least someof thesemeaningsmay have been derived
fromeconomicuses of the spruce,oftenemployedfor shingles.
Both sememeshave previouslybeen connectedwiththe adjective
i, e hollë'thin,fine9by Çabej (StudimeFilologjike,1962.4: 112)
see also the remarksby Hamp in Evidenceforlaryngeals(1965:
130).
Wörterbuch,
2) Lambertz (Albanisch-deutsches
1954) defineshalë as 'Fichte,
spruce*.Jaan Puhvel (per litteram)has called my attention to the further
possibilitythat the Hittite animate tree-name G1&alanza(n)-may be cognate
with Lat alnus 'alder*. If cognate, the absence of h- in Hittite is welcome
confirmationof *A2 in halë. alanza(n)- was formerlycompared to Gk ελάτη.
Such an etymologyprovides no justificationfor the affricate-z-. The word
may reflect *A2el(k)sno~metathesized to *alnteo-with epenthetic -t- or
*k assimilated to t or simply *A2eln-so-,
a reduced formof *aleno-(an alternate source of Lat alnus) with an -$o-suffix.
The correctgloss for'silver fir,
Abies cephalonica* (the probable referentof ελάτη)seems to be G1&tanãuwhich GünterNeumann plausibly equated withNHG Tanne (KZ, 77 [1961];
77-8). Professor Puhvel furtherinformsme that alders still grow along
stream beds in the Anatolian plateau.
8) This and several other instances of apparent Proto-Indo-European
polysemyor polylogycould well be ascribed to the change in the arboreal
environmentthat coincided with the climatic change from Atlantic to
the SubBoreal around 3300 BC.

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

Albanian corrigendato Friedrich's Proto-Indo-Europeantrees

305

*ar-: arrë'walnut,Juglansregia9[arre],arrë is a substantized
no-adjective,a formationcommonto both Albanian and Latin.
Contraryto what Friedrichsays (78) -rr-,a trilledliquid, is not
thereflexof*-sr-,whichin factbecomesa flap,-r-,ci dore'hand*<
rrarisesfrom*rn,eg marrΊ seize' < *marnioh.
Forms
*§heesrek]
with initial aspiration,harre,appear to be unetymological
; a
dialect study would reveal whetherthese like other inorganic
corrections.
aspirationsare hyper
*wyg:vidh,-i 'elm, Ulmus campestris'[vith],Pekmezi (Grammatikder albanesischen
Sprache,1908: 22) cites a nasalized Gheg
modern
form,v{o (in
vidh).Nasals beforebothoriginal
orthography
sibilantsand thosearisingfromdorsalsare subjectto earlierloss
nasals hence it is difficult
to confirm
than otherpreconsonantal
the nasalization Pekmezi reports.Cimochowski(Le dialect de
Dushmani,1951:48) records/viö/withoutnasalizationbut also
If the nasal vowel is originalthis would be an
/mis/< *memso-.
Alb vidh and Slavic (Russ vjaz) and Baltic
between
agreement
(Lith vinkëna)< *uin§-againstGermanic(OE wice)and Iranian
(Kurd viz 'elm' and Osseticwis-qäd'maple') < *ui$-o-,
'lumber'[Vende],
*lenTã: lend(e)'acorn',Us 'oak, Quercus',Vende
Friedrichnotes but stronglyquestionsincludingVendewith the
root *gwelH-'acorn' withwhichit is not related; *gwelH-would
give Albanian *zel-; *gwlH-Albanian *gal- or *gla-,lend(e) representsthe e-gradeseen in OE lind 'linden',Lith lentà'board'.
Us is a substantizedto-adjectivewith*-tt-becoming-s- and ♦-enThis same
becomingi beforea spirant,cfmish'meat' < *memso-.
in
: : uent-to-.
is
seen
:
:
*uent<
vend(e) vis(e) 'place'
relationship
in
is an o-gradeas
Vende
OCS iqJtü
'linden'; Gk ελάτη'pine?' is a
The wide arrayof ablaut formsindicates
zero-grade< *(?)lrit-eh.
one of the most commonand economicallyimportanttrees in
and the shiftto 'oak' in Albanian
the Indo-Europeanenvironment
to the mostavailand 'pine' in GreekreflectBalkan substitutions
1964.2:
able replacement.
21) suggested
Çabej (StudimeFilologjike,
a relationto Lat lens 'lentil' forlende'acorn' withoutjustifying
Us.
the semanticsof Vendeor considering
*os-: ah, -u 'beech, Fagus' [ah]. The semanticmatch is with
Gk όξύη'beech'. Phonologicallyah matchesOE sesc'ash' < PG
*askaz and Makedonianαξος'ϋλη
(wood,tree) < PIE *xesJco-,
*grobh-:shkozë'beech,hornbeam,Carpinusbetulus'[shkozë(î)].
Friedrichhas reservationsabout the equation proposedby Jokl
20·

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

306

Martin
E. Huld

{Wörterund Sachen, 12 [1929]: 63-91) relatingshkozëto Baltic
hornbeamterms.Beside the attestedlong-oforms,Joklproposed
forAlbanian a long-egrade - a formation
that is productivein
Albanian,cf Jokl (IF, 37 [1916]: 190-222) plus the collective
suffix-zë. Since Jokl believedthat *sk-became shk beforeback
vowels but ç beforefrontvowels,the loss of the first*r in *smust be carefullytimed afterthe loweringof original
Krêb-r-zë
In
*ee to o.
Friedrich'sskepticismis
lightof these requirements,
justified.Consideringthe hardness of hornbeamwood, shkozë
could reflecta lengthenedgrade of the root kaar-chard5with an
*sV- prefixand -zë collective.Contaminationfrom*kar-'hard'
seems to have occuredin Lat carpinusas well.
*bhïgo-: bungë,also bung,-u*chestnutoak, Quercussessiliflora*
[bunge(??)]. Despite the strangeappearanceof bungëthereis no
reasonto expressdoubt. Semanticallybungëagreeswiththe shift
of meaningalso seen in Gk φηγός'oak'4). Formallybungëis a
substantizedno-adjective,*bheA§-neh
; *$ is regularly
depalatalized
cf
before consonantalrésonants, Hamp KZ, 76 [1960: 275-80.
The sequence *gn then metathesizedto -ng-s).While PIE *[aa]
as u before
normallybecomesο in stressedsyllables,it is reflected
c
<
nasals, eg shtrungëcattle-pen(for milking)' *[straag-neîi]?
see also Hamp
originallya slaughterpen,cfLat strãgês'slaughter9
20 [1978]: 115-19.Çabej's suggestion
Balkanskoezikoznanie
(Buletin
i Universitetit
shtetëror
të Tiranës,ser. shk. shq., 1960.4: 77-78)
that bungëmaybe relatedto Olr bun'trunk(ofthe body)5is both
and semanticallyimprobable.
morphologically
Cornusmas' (θαηε(? )]. Friedrich's
*k(e)rn-:thane'cornelcherry,
doubts are well founded; *-rn-should give -rr-not -n- and th
reflects*lc not *Jfe<»>.
Çabej (StudimeFilologjike,1966.3:53-52)
4) Because the beech (Fagus silvatica)is commonin the Balkans in factthe secondmostprominent
memberof Albanianforests(Keefeet
al, Area handbook
forAlbania, 1971 42-34) - and Greekand Albanian
designatethistreeby'ash' - bothtreesproducehard,durablewoodsuitable
fortools,handlesetc, it is clear fromthe Greekand Albaniansemantics
that *bheAgoto the Northneverdesignated'beech',a meaningconfined
west innovatingarea, but some othermemberof the Fagaceae familyto
in thenorthbutthechestnut
whichthebeechborethegreatestresemblance
oak in the Balkans.
5) The processthat looks like metathesiscould actuallybe the assimilationof *-gn-to a homorganic
nasal -y-,cf *-pn-> -m-in gjumë'sleep' <
This marginalphonemewouldhave beenidenticalto theprevelar
*8wp-no-.
allophoneof /n/and acquireda laterepentheticjgj.

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

's Proto-Indo-European
to Friedrich
trees
Albaniancorrigenda

307

and hesitatingly
endorsedPedersen'sequarejectedthisetymology
tion(KZ, 36 [1900]: 332) withOlr congan'horn'.Whilethe phonology of this equation is probablycorrect,derivingthane'cornel
from*lcon§-nã'horn' seems to be more a pun based on
cherry9
Lat cornus: : cornu'cherry: : horn5than an etymology.Further
is hazardous;but as a guess,if Lat sorbum'serviceetymologizing
- it is not clear whether
can
be
derivedfrom*sosdhom
berry9
the rhotacismin mergo'wash9and orbis'sphere9(cf Lith mazgoti
and Welsh oddf)or loss as in nidus 'nest9is the regularoutcome
of *s beforevoiced (aspirated)stops, thanemay be froma substantized*sosdh-neh
; *s becomes*ts,the antecedentof jdj (orthoanother
*s, eg than 'it is dried9< *sausnit,
graphicth) before
the cluster*dhnwould regularlybecome a non-rhotacizing
-N-,
hence *sosdh-neh
> téas-No> thane.
*perkw-:Perëndi,Perëndija'god (head)9[Perëndija(?)]. Again
thereis littlereason within
Friedrich'sdoubts are well-founded;
Albanian to connectthis word with 'oak9. The phonologicalreis slightand a relationcannotbe justified
semblanceto *perkwFilologijke,1965.1:
Çabej (Studime
throughregularsound-changes.
19-20)notedthatthemeaning'emperor9
mayhave beentheearlier
The shape of the wordlooks suspiciouslynon-native
signification.
to me and my own,hardlytestable,guessis that it is a loan from
Slavic Perunü divü with the latter word cognate
pre-Christian
with Lith diëvasand subsequentlyousted by Iranian bog-*).
an errorforbVetezs.
5-birch:bletëzë
[blétsze,
Meyer'sinterpretation
of a formoccuringonly in Blanchus' DictionariumLatino-Epiroticum,1635]. The formbletezeglosses betulaefromwhichit is
borrowed.The apocopationof late Latin -Cul~to -Cl- is well attestedin Albanian,eg ungj 'uncle9< *avunclo< avunculus;thus
betulae> *betl(a)eand by metathesis*blete-with a common-zë
suffix,'birches9.Çabej's rejectionof this etymologyin favor of
an obscurerelationwith mëllëzëor mbëdhezë
'hornbeam,Ostrya
i
Universitetit
shtetëror

Tiranës,
ser.shk.shq.,
carpinus9(Buletin
1960.4: 62) is doubtful.The formerat least may itselfbe a loan
fromLat millettia.
verr,-i 'alder, Alnus9[verre,versprobablythe
3-alder/8-elder:
same wordwithoriginalrr writtenf in the philologicaltranscripe) Old Russiandivü'demon'is thoughtto be a loan fromIraniandev-,
but thiswordcould be a nativeformservingas a caique on the Iranian
semantics.

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions

MartinE. Huld

308

tion and the macron later omitted)].Friedrichglosses this as
'white poplar9afterMeyer (1891:467). verrand verreare from
c
*ufne-,exactlymatchingOlr fernand Welsh gwernalder'.
12-hazel: lajthí, -u; lejthi,-i 'hazel, Corylusavellana5 [lethi,
laithi],Çabej (StudimeFïlologjike,1964.2: 113) suggesteda relation
to lak 'sling,tendon'afterthe slenderbranchesof the hazel but
the phonologyof such an equation is imprecise.The diphthongs
cf i, e drejtë'straight3< Lat
aj and ej pointto a late borrowing,
diréctum
is unclear.
; but the sourceof such a borrowing
Friedrichsuggestedthat Albanianshowedmorenotable agreementof tree names with Greekthan with otherbranches(165).
A closerexaminationshowsonlythreeexclusivelyshareditemsthe semanticchangeseenin bothah and bungë(Gk όξύηand φηγός)
and pisë and Gk πίσσα'tar'. This is hardlymoresignificant,
given
our presentstate of knowledge,than the numberof shareditems
with Baltic-Slavicand Germanic(vidh and the full-gradelëndë
againstGreekzero-grade)or withLatin (halë,pishë,and perhaps
thane).Perhaps as Friedrichsuggestedthe Greek and Albanian
agreements
mayresultfromthe commonBalkano-Aegeanenvironment;however,Albanianalso sharesa termwithGreekand each
of these othertwo groups- molle,Gk μήλον,Lat mãlum'apple'
and arrë,Gk αρνα,Russ orèx 'walnut'. Such a distribution
may
indicatean intermediate
positionforAlbanian betweenMeillet's
Northwest,
especiallyBaltic-Slavic,and a southerngroupofwhich
Greekwas theprinciplemember.Thisinterpretation
agrees,though
not in all details,withthosepresentedelsewhere,eg Pisani (L'albanais et les autres langues indo-européenes,
Annuairede Vlnstitutde philologie
ethistoireorientales
etslaves,10 [1950]: 519-38),
des indogermanischen
Porzig (Die Gliederung
1954),
Sprachgebiets,
and Huld (KZ, 92 [1978]: 297-99)7).
1605So. CordovaSt.
Alhambra,CA91801

Martin E. Huld

doesnotdenythevalidityofa widerIndo-European
7) ThoughFriedrich
contextthat linksAlbanianespeciallywith Baltic-Slavic,he informsme
to the threesharedin(perlitteram)that he stillfeels'a real significance
novationsin question. . .". The difference
in our views lies preciselyin
the importanceattachedto these innovationsratherthan acceptingor
rejectingetymological
equations.

This content downloaded from 158.121.247.60 on Sun, 13 Dec 2015 09:39:28 UTC
All use subject to JSTOR Terms and Conditions


Related documents


PDF Document albanian pie to friedrich
PDF Document albanian phonics
PDF Document palatal resonant messapic
PDF Document health and nutrition in pre columbian america
PDF Document living color
PDF Document 650647


Related keywords