C H J VAN DER MERWE
at Stellenbosch University
Jan Joosten (Oxford), Meir Malul (Haifa), Cynthia Miller-Naudé
(Bloemfontein), Jacobus Naudé (Bloemfontein), Herbert Niehr
(Tübingen), Hermann-Josef Stipp (München), Ernst Wendland
(Lusaka), Arie van der Kooij (Leiden)
Department of Ancient Studies
I CORNELIUS G R KOTZ
C H J VAN DER MERWE
The Journal of Northwest Semitic Languages
(ISSN 0259-0131) is published half-yearly
JNSL is an accredited South African journal. It publishes peer reviewed
research articles on the Ancient Near East. As part of the peer review
policy all contributions are refereed before publication by scholars who
are recognised as experts in the particular field of study.
Contributions and books for review should be sent to
The Editor: JNSL
Department of Ancient Studies
Private Bag X1, Matieland, ZA-7602
Subscriptions should be sent to the same address but marked as
Department of Ancient Studies, Stellenbosch University, Stellenbosch,
Articles submitted for publication must be according to the house rules
on the homepage
JNSL homepage (house rules, contents, subscription)
TO ORDER: Journal of Northwest Semitic Languages
Send an e-mail to Ms L C Swanepoel (email@example.com)
For further subscription information: e-mail Ms L C Swanepoel (firstname.lastname@example.org)
Per Invoice $ 70.00
Booksellers - 30%
Journal of Northwest Semitic Languages 43/2 (2017), pp. 97-123
David Van Acker1 (University of Leuven)
, AN ETYMOLOGICAL AND SEMANTIC
is a dubious word in the Hebrew Bible. It has been commonly interpreted as a
compound noun - , translated with “shadow of death” or an abstract noun
from the Semitic stem ṣlm-II, “darkness”. However, both readings are cumbersome: a
translation “shadow of death” fits badly in most contexts of , while the mere
existence of ṣlm-II in Northwest Semitic is problematic. With some new evidence from
Ugaritic research, I will argue that the - etymology is to be preferred. However,
the translation should indeed be centred around the concept of ‘darkness’. Using
collocation analysis and insights from cognitive semantics, I will make a semantic
analysis of the term, charting all its meaning aspects. Thus I will demonstrate how
has a conceptual structure including the notions of ‘darkness’, ‘terror (of death)’ and
‘locality’. These notions can be explained with a - reading and less so with a
For over a century, scholars have been debating the meaning of the word
. It occurs only 18 times in the Hebrew Bible, but is present in some
very well-known passages including Ps 23:4, Isa 9:1 and several instances
in the book Job. Generally, the word is interpreted in one of two ways: (1)
The Masoretic pointing of is followed, reading as a compound
of and , generally translated with “shadow of death”. (2) The
Masoretic pointing is considered a folk etymology and it is argued that the
consonantal form should be read as or , an abstract noun
derived from the stem ṣlm-II (to be dark), otherwise unknown in Hebrew,
but often attested in other Semitic languages. Although a substantial
amount of arguments has been adduced for both interpretations, translators
and interpretators alike are still divided on the matter.2 Unfortunately, no
1 PhD Fellow of the Research Foundation – Flanders.
2 In the TDOT Niehr reads even though he acknowledges nuances which
are more evident from the reading (1977). In the TLOT, on the other hand,
Price observes the same complex semantic field and he categorises under
the stem -III along with (1997). This is the same approach as the one used
98 DAVID VAN ACKER
scholar thus far has succeeded in bringing together all the available
evidence and weighing the arguments against each other.3 Moreover, all
scholars, except Cohen, have chosen a philological or etymological
approach, while I hypothesize that a semantic approach is preferable in this
The earliest interpretations of are unanimous: it should be interpreted
as a compound noun of and . Hence it is translated σκιὰ θανάτου in
the LXX4 and also the Targum and the Peshitta follow this interpretation
(Thomas 1962:192). The Vulgate takes over this interpretation, translating
by Clines in his dictionary, however he mentions the other possible readings as
alternative interpretations (1993: III). In his contribution to the Word Biblical
Commentary, however, Clines clearly advocates the - interpretation and
urges translators to include “death” in the translation, albeit not as a physical
reference to the underworld, but a superlative (1989:223), as Thomas argued
3 The arguments of early scholars have been discussed in full and are expanded
on by more modern scholars. But modern scholars have neglected the possible
evidence of Ugaritic research (Barr 1974b; Michel 1984; Cohen 1996),
diminishing the evidence from the LXX (Cohen 1996), leaving earlier questions
unanswered (Mazzini 1999) or failing to formulate a clear conclusion (Stone
2006). Some of these (at least Mazzini and Stone) fail to mention notable
previous studies entirely. A number of other contributions mention , but do
not add to the discussion, because they follow interpretations that suit their needs
(Tromp 1969:140-144; Mittmann 1980:9-10; Jones 2009:133-134; Hankins
2013:214 n. 8; Balogh 2014:532; Whitley 2015:129 n. 5).
4 12 out of 18 times (Pss 23:4, 44:20, 107:10, 14; Job 3:5, 12:22, 24:17 [2x], 28:3;
Isa 9:1; Jer 13:16; Amos 5:8). Other translations in the LXX are γνοφερός (Job
10:21) σκιὰ (Job 16:16) ᾅδης (Job 38:17) and ἄκαρπος (Jer 2:6), all occurring
only once. Twice the word has remained untranslated (Job 10:22, 34:22). In the
first of these two, the phrase has been translated as εἰς
γῆν σκότους αἰωνίου and in the second instance has been
translated as οὐδὲ ἔσται τόπος.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 99
it as umbra mortis5 and ultimately the Masoretic pointing also indicates a
In more recent times the word has been revisited and reinterpreted from
various perspectives. On the basis of linguistic achievements such as
methods for comparative linguistics as well as the discovery and
decipherment of cuneiform texts and subsequent better understanding of
the Semitic language family, scholars have proposed new explanations for
. It was, for example, argued that the traditional - interpretation
is based on a folk-etymology, once the original stem ( * ) and its meaning
had become unknown. The original vocalization of should in this
case have been , an abstract noun of the stem ṣlm-II, or it could have
the rarer - abstract ending to justify the Masoretic change in vocalization
(Cohen 1996:306-308; Jones 2009:133-134).7 The same root is well known
in Akkadian (ṣalâmu), Arabic (ملظ, ẓlm) and Ethiopic (ṣalma).8 The
following arguments have been given to justify such an interpretation:
1. The root ṣlm-II (to be dark) is a well-known, widespread stem in the
family of Semitic languages (Thomas 1962:193-194; Driver & Gray
1964:II 18-19; Eybers 1972:24ff.; Clines 1974:17ff.; Bordreuil &
Caquot 1980:348; Cohen 1996:288-289; Mazzini 1999:79).
5 Or tenebrae (darkness, Job 24:17b; 38:17; Amos 5:8), imago mortis (image of
[the] death, Jer 2:6) – showing the first signs of confusion on the term –
opertam mortis caligine (the hiding of the darkness of [the] death, Job 10:21),
and caligare (to be dark, Job 16:16).
6 It must, however, be noted that the vocalisation of has been changed to
(with pataḥ), rather than its regular construct form (with ṣere). This change is
not without parallel however, a very similar word (from original *qinn, as is
> *ṣill) has the construct form (with pataḥ) found in Deut 22:6.
7 This vocalisation explains how the Masoretes could vocalise the consonants as
. If one would reconstruct them as -, this change cannot, as Barr
(1974b:52) points out, be explained. This - ending is explained by Cohen with
a parallel with another word meaning darkness that also has this abstract ending
in one case: ) ( in Isa 59:9 (Cohen 1996:306-308). However, this might
just as well be a poetic assimilation with the preceding word ) ( which is a
plural feminine form. The form in Isaiah should therefore be considered as an
occasional deviation from the ending - rather than as a structural form. This
can hardly be an argument to accept the structural use of throughout the
8 First enumerated in full by Eybers (1972:29).
100 DAVID VAN ACKER
2. Shadow has positive connotations in Near Eastern cultures (Thomas
1962:195; Driver & Gray 1964:II 18-19; Mazzini 1999:80), therefore
its connection with death is counter-intuitive and its use in dark and
pejorative contexts as is unlikely.
3. Compound nouns are rarely attested in Biblical Hebrew (Driver &
Gray 1964:II 18-19).
4. The translation “shadow of (the) death” is unnecessary and untenable
in some verses where is found, whereas “darkness” fits better
in all the contexts where it is encountered (Driver & Gray 1964:II
18-19; Cohen 1996:301-303; Stone 2006).
On the other hand, many scholars have defended the traditional reading,
giving the following arguments:
1. The root ṣlm-II is not attested to in Hebrew, while in the languages
where it occurs, it occurs relatively frequently (Michel 1984:8),
moreover, when in Hebrew the abstractum - is used, there are
usually more derivatives from the same root (Thomas 1962:195).
Therefore, if ṣlm-II were attested in Biblical Hebrew we should
expect to see more of the lexeme in the Hebrew Bible.
2. The positive meaning of shadow can be altered by the mere co-
occurrence with death, or alternatively if is read as “shelter”, it
can refer to the dwelling of Death or Mot, being the underworld
(Thomas 1962:195; Barr 1974b:54; Michel 1984:7, 12).9
3. Compounds do exist in Hebrew, especially in person or place names
(Barr 1974b:54; Michel 1984:7, 10-12).10
4. is a complex word where an univocal translation is not possible
(Barr 1974b:50; Michel 1984:12), furthermore, one must not neglect
the versatility of Hebrew poetry, nor the nuances of meaning (Michel
5. The LXX translates with σκιὰ θανάτου11 and most passages
containing are considered to be written around the same time
9 It must be added that twice in the Hebrew Bible is connected with the darkness
of night, showing that it is indeed possible to deviate a little from the meaning
of ‘(positive) shade’ – Cant 2:17 and 4:6;
– Until the
day breaks and the shadows flee (NIV).
10 This enables us to understand as an ancient place name for the shelter of
Mot, being re-interpreted later (I will elaborate on this below).
11 Many scholars have proposed the LXX translation of σκιὰ θανάτου as an
argument in favor of - . There is no reason for Cohen to diminish the value
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 101
as the composition of the LXX (Thomas 1962:195; Barr 1974b:51-
52; Michel 1984:8).12
For the sake of completeness, I need to mention two other interpretations
1. Thomas interprets as a superlative of , using as a
superlative modifier and thus translates it with “deep darkness”
2. Haupt interprets as an intensive plural, reading and to
be translated as “deep darkness” (1914:221-222).
Examining the arguments above, it becomes clear that most of the
arguments for the ṣlm-II reading can be countered by arguments for the
- reading. However, it remains true that the latter’s proposed
translation of “shadow of death” is not an appropriate reading in many of
the instances of . The etymological basis of a ṣlm-II stem in Hebrew
(or Northwest Semitic), on the other hand, is questionable. In the following
of the LXX or to claim that evaluating it is not the task of a ‘biblical philologist’
(Cohen 1996:301) – whose job would it be otherwise? Contrary to what Cohen
claims, the value of the LXX is not limited to confirming the consonantal
Vorlage of the translators. It also provides an interpretation of this consonantal
text and gives an insight in the contemporary tradition of its (oral) vocalisation
12 However, the dating of the occurrences of is no unambiguous matter. It is
possible that – at least for some of its occurrences – there is an older tradition
behind the text. Therefore some scholars have been tempted to claim that the
history of includes an original form derived from the root ṣlm-II, which
has been forgotten and reinterpreted with - before the composition of the
LXX. This means that by the time of the final redaction of the Hebrew Bible, the
redactors themselves understood the form as - , justifying the LXX
translation. Therefore we cannot rule out that the LXX presents a reinterpretation
of the consonants . It does, however, show that the - interpretation
predates the final Masoretic Text with more than 1000 years.
13 This reading, however, cannot be maintained. It is certain that is a positive
term and has nothing to do with ‘darkness’ albeit that in collocation with ‘death’,
this positive ‘shade’ or ‘shelter’ can be corrupted to the ‘shelter of death’
implying ‘darkness’ with a pejorative connotation. However, when the ‘death’
element in this compound should be a grammatical substitute for making a
superlative, there is no way to explain the semantic shift from shade with a
positive connotation to darkness with a negative connotation.
102 DAVID VAN ACKER
paragraph I will evaluate some of the etymological arguments for and
against a ṣlm-II reading and add some new elements to the discussion. I will
show that the most likely etymological reading is - . In the third
paragraph I will discuss the meaning of building on cognitive
semantics, rather than on etymological grounds. This will show that
features a prominent aspect of ‘darkness’, amended with aspects of ‘terror
(of death)’ and ‘location’. In the conclusion I will eventually show that
given this semantic structure and the evidence against a ṣlm-II reading,
, must indeed go back to a compound noun of and . This,
however, does not imply that the meaning ‘shadow of death’ must be
preferred also. In the third part I will make clear that the meaning ‘darkness’
is in some cases appropriate, also even with a - etymology. In such
cases a translation of “deadly darkness” or “dark place” may be justifiable
– depending on the context.
3. EVALUATION OF THE ETYMOLOGICAL ARGUMENTS
3.1 ṣlm-II in Northwest-Semitic
The basis of the etymological discussion is rightly condensed by Mazzini
(1999:80) to the supposed existence of the stem ṣlm-II in Hebrew and
Northwest Semitic. With ṣlm-II meaning ‘to be dark’ and ‘darkness’ being
a suitable translation for throughout all its occurrences, more so than
‘shadow of death’ (Cohen 1996; Stone 2006), such an etymology seems
reasonable. However, there must be more arguments to accept this
etymology besides that it provides a better translation. Translations often
disregard the broad spectrum of meaning behind an original word (Barr
1974a:15-16)14 which I will call the conceptual structure (Riemer
2010:239). So if a translation as “shadow of death” seems unfitting to
Cohen and Stone, they might first want to do a semantic study of .
This is what Cohen attempts to do, but I disagree with his methodology (cf.
section 4.4). Moreover, even though ‘darkness’ should appear as an
important aspect of , this does not necessarily mean that a -
etymology is wrong, as I will show later on. Thus there must be good
14 And I will show later (cf. section 4.4) that the necessary nuances to add to the
broad meaning of ‘darkness’ are ‘terror (of death)’ and ‘location (of the
underworld)’, which are all obvious from the - interpretation.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 103
reasons to abandon the Masoretic pointing in favour of an otherwise
unknown stem ṣlm-II15 stem in Northwest Semitic.
3.2 Ugaritic Evidence
3.2.1 The Texts
If ṣlm-II were attested only in non-Northwest Semitic languages such as
Akkadian and Arabic, we could easily abandon the ṣlm-II reasoning here.
However, the corpus of Ugaritic texts has shown four cases where the
supposed stem ṣlm-II might be attested. Some of these texts are mentioned
briefly by Barr, Michel and Cohen, but none of these scholars provides an
in-depth analysis of the texts. Mazzini (1999) was the first to discuss three
of these texts in full.16
The most recently found and most intriguing text adding to this
discussion is KTU 1.169. It was published in 1980 by Bordreuil & Caquot
and thus not yet known by most scholars who wrote on .
7. (…) bmrmt on the heights
͗phm in the watery (plains), in the darkness, in
the sanctuary, here
9. kšpm.dbbm.ygrš.ḥrn did the magicians conjure. Horon hunted
10. ḥbrm.wġlm.dʿtm.lk his companions, the young man, the
acquaintances are in your favour.17
In another text – KTU 1.161 the same combination of letters can be found:18
spr.dbḥ.ẓlm Instructions for the Statue-Offering
qrim.rpi.a[rṣ] You will call Rapi’u of the Under[world]
15 The fact that ṣlm-II is elsewhere unattested in Hebrew is in itself no argument
against its realisation in . However, adding to this Nöldeke’s claim – as
cited by Thomas – that abstracta on - (or - ) do usually have derivatives in
Biblical Hebrew (Thomas 1962:195), the or etymology, becomes
16 Mazzini reacts in part to a treatment of the Ugaritic texts by Clines (1974:21),
but the latest discoveries, important for our current discussion, were yet
unknown to him.
17 Author’s translation of the French translation by Bordreuil & Caquot (1980:350).
18 However, in this case it is clear that the stem is not ṣlm-II. Nevertheless I present
this case because it puts the other passages in perspective, as they also could
have the same meaning as in this case.
104 DAVID VAN ACKER
qbitm.qbṣ.d[dn] You will fetch the gathering of Di[danus]19
The other two texts are parallel texts from the Baʿal cycle, KTU 1.8:7-9 and
KTU 1.4.7:54-56. Baʿal is sending messengers to Mot:
7. (...) bn.ġlmt 54. (…) .b<n>ġlmt in the darkness
8. ʿmm.ym.bn.ẓlm[t] 55. [ʿmm.]ym.bn.ẓlmt.r is involved the
sea, in the
9. rmt.prʿt (…) 56. [mt.prʿ]t (…) the primeval
The first text cited, (KTU 1.169), is rather fragmentary and its context and
nature are not certain. It can be either a mythological or a magical-
incantation text (Mazzini 1999:80). Both Bordreuil & Caquot and Mazzini
opt to read a form of ṣlm-II here,21 otherwise unattested to in Ugaritic.
However, ẓlm an sich may be more than merely a form of ṣlm-II. In fact,
Del Olmo Lete & Sanmartín (2003:1004) read this word as the plural of the
well-known noun ẓl (shade/shelter, equivalent to Hebrew ). The
translation would then turn out to be:
“on heights, in the watery (plains), in the shade of the / in the
This interpretation was suggested by Del Olmo Lete & Sanmartín
(2003:1003-1004), creating a link between bẓlm and the following bqdš,
which is not present in the original translation.
The second text, (KTU 1.161), is not included in any discussion on
thus far, although it is the only other example of the three letters ẓ-l-m
appearing together in Ugaritic. However, the same discussion as given
above applies here. This word has been interpreted as a plural of ẓl or as a
form of ṣlm (Dietrich & Loretz 1983:18).
The other two texts, (KTU 1.8 and 1.4.7), hardly represent any
convincing evidence. In his recent edition of the Baʿal epic, Mark Smith
qualifies this passage as “very difficult” (Smith & Pitard 2009:657) and he
places many question marks in his translation.22 In his interpretation of the
19 Author’s translation of the German translation by Dietrich & Loretz (1983:23).
20 Translation provided by Mazzini (1999:83) in his treatment of .
21 Called ẓlm by Mazzini.
22 54. (…) b<n.>ǵlmt (…) so(ns) of the Lass (?),
55. [ʿmm.]ym.bn.ẓlmt.r [kinsmen of ]Day (?), sons of ẓlmt, the
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 105
second fragment (KTU 1.8) he supports Pardee’s claim that the fragment in
1.8 should actually be placed at the beginning of 1.3.6 (Pardee 2009). This
moves one of the two parallel texts away from the context of Mot and
consequently makes an interpretation closely linked with the underworld
unlikely (Smith & Pitard 2009:694). But – as Smith admits – this adds no
more clarity to the passage (Smith & Pitard 2009:695). In Smith’s
translation ẓlmt remains untranslated but it is suggested that it be some kind
of name and an epithet for Gpn wUgr. In that case bn is being translated as
“sons of”, rather than bn, “between” or an alternate spelling of b, “in”.23
The advantage of such a translation is that it provides a generic
interpretation that is independent of the context of the passage.
Every other interpreter of these texts has made the connection between
ẓlmt and , and has connected the term with ṣlm-II. However, this
results in a circular reasoning. Neither word has an unambiguous meaning
nor are they unambiguously connected to ṣlm-II. Therefore ẓlmt cannot rely
on for its interpretation, nor vice versa.
I consider the second cited text (KTU 1.161) as key to understanding the
first text (KTU 1.169). Here we find, quite unambiguously, the letters
ẓ-l-m and there is little discussion about reading this as “statue”. If we
return with this knowledge to the first text, its meaning and structure
become more evident. Rather than creating a hapax with a ṣlm-II root in
Ugaritic resulting in the proposed translation “on heights, in the watery
(plains), in the darkness in the sanctuary…” we can read “on heights, in the
watery (plains), on the statue in the sanctuary…”. This does create a
conceptual link between the two elements in the second part of the phrase
(statue and sanctuary, both important in worship), as is already present in
the first part (heights and plains, both geographical elements). Thus I see in
this a rhetorical merism, meaning ‘from high to low’ and ‘from the
(smallest) statue/image/idol to the (largest) sanctuary’24. Moreover this
56. [mt.prʿt.] (…) [exalted princess] (…)
23 Called respectively bn (I), bn (II) and bn (III) by Del Olmo Lete & Sanmartín
24 In that case the word ẓlm would not refer to the main statue of the deity within
the sanctuary, but rather the multitude of smaller statues owned privately and
used for a number of religious practices including incantations. This use of the
word ẓlm/ṣlm in Ugaritic can hardly be verified due to the limited attestation of
the word. However, it is clear from Akkadian that ṣalmu can be used to refer to
106 DAVID VAN ACKER
reading avoids the awkward interpretation of the second preposition b-
created by Del Olmo Lete-Sanmartín (2003:1004) in their translation “on
heights, in the watery (plains), in the shade of the / in the sanctuary…”.
The case of KTU 1.8 and 1.4.7 is quite different; since here we have the
final taw added to ẓlm. As such these occurrences are most similar to .
As discussed above, Smith leaves this word untranslated and regards it as a
proper name or an epithet of Gpn wUgr. Other authors identify ẓlmt with
darkness. However, a geographical description of the location from where
or to which Gpn wUgr is commissioned, is still also possible, as well as
ẓlmt being a compound of ẓl and mt, either referring to the underworld, the
dark ends of the earth or even to Gpn wUgr originating (bn) from the
vicinity (ẓl) of Mot (mt). Given our current understanding of ẓlmt and
a connection with ‘darkness’ has to be made in any case. Thus Mot might
well be called to the scene,25 if only in the epithetic address of Gpn wUgr.
Whatever the context and meaning of this passage might be however, the
meaning of ẓlmt remains uncertain.
The evidence adduced for a Northwest-Semitic stem ṣlm-II can be
explained more easily without taking recourse to the ṣlm-II etymology. As
a consequence, serious doubts arise with regard to the existence of ṣlm-II
in Northwest-Semitic or Biblical Hebrew in particular. And without this
evidence, the hypothesis can hardly be supported.
Contrarily, examining the individual cases of ẓlmt in Ugaritic (KTU 1.8
and 1.4.7), it appears that this term is highly ambiguous. It might be
connected with notions of darkness and death, and a location in heaven.
This leaves room to interpret ẓlmt and consequently as a compound
place name – a thing quite common in West Semitic languages – composed
of ẓl (here meaning ‘shelter’, ‘vicinity’ or even ‘cover’ rather than
‘shadow’) and mt (being the personification of death, the well-known
Ugaritic deity Mot, rather than death as an abstract notion), meaning Shelter
of Mot as a place name referring to the underworld. Unfortunately the few
attestations of ẓlmt in Ugaritic are insufficient to unambiguously interpret
this term as either derived from ṣlm-II or ẓl and mt. However, it seems
small figurines, household deities or any other image of a deity, especially in a
context of incantation (CAD:ṣalmu section d). Moreover, archeological evidence
points out that such small figurines were also common in Ugarit (Yon 1997:88,
25 Smith, however, thinks this is unlikely (Smith & Pitard 2009:396).
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 107
unlikely that ṣlm-II is to be read here. If we would follow this hypothesis
further it is not unlikely that such a place name has been used in the Hebrew
Bible26 in a more abstract way generalising it to the underworld or even the
kind of darkness associated with such a place.
Thus these Ugaritic texts cast doubt on a reading, but they do not
provide sufficient proof for a - reading either. Therefore it is to a
semantic study of that we must turn in order to come any closer to a
more justifiable interpretation of the expression.
4. CONCEPTUAL STRUCTURE
I will not add an in-depth analysis of the other arguments for a -
reading here. I have enlisted the arguments briefly in the beginning of this
article and they are adequately described by the cited authors. My remarks
on these arguments can be found in the subsequent footnotes. By now I
hope to have made it clear that an interpretation of as a compound
noun of - is more evident than interpreting it as or .
However, all this stands or falls with the resulting meaning, for it has
already been stated that the “shadow of death” translation seems not to fit
for all occurrences of . Let us therefore establish a conceptual
structure (i.e. the mental state behind the used concept) based on all the
cases of (through an analysis of its context and collocations27), and
see how this can be explained by a - etymology.
The conceptual structure of a word is presented in our minds as an
idealized cognitive model (ICM) of the described word (Riemer 2010:240-
241). This includes all “implicit knowledge we have about the objects,
26 The fact that this produces a Hebrew compound noun can be explained in two
ways: (1) It either is a loanword in Biblical Hebrew. The Hebrew word
(temple), is an example of a compound loanword, derived from the Akkadian
compound ekallu (É.GAL, temple, lit. great house); this loanword may
subsequently have underwent a change in meaning. The Hebrew word
(officer) from the Akkadian ṭupšarru (scribe) exemplifies the possible shift in
meaning of a loanword. (2) Alternatively at some stage Hebrew also used a
similar place name which has over time evolved into a more abstract noun.
Compound place names are very common in Biblical Hebrew, some even
contain or . For example the name (lit. enclosure of death), this
provides a parallel with for it seems to indicate some kind of location
similar to and ḥṣr is also attested in Ugaritic; or the name (lit. the
shadow of Paḥad). For more examples see Michel (1984:10-11).
27 For an introductory treatment of this methodology see Riemer (2010:386-389).
108 DAVID VAN ACKER
relations and processes named in language”. Thus everything there is to
know about is included in this ICM. This pool of knowledge in turn
influences the way people understand its use in different contexts. Hence in
every context different aspects of the same ICM appear more prominent
than others, while the ICM remains the same (Riemer 2010:250-251). It is
thus our task to determine which aspects are included in this pool and which
are generally more prominent than others.
4.1 The ‘Darkness’ Aspect
The initial reason for scholars to reject the meaning of ‘shadow of death’
for was that it did not fit many contexts of the word well. ‘Darkness’
seemed more probable. This can be seen in the many instances where
is parallel with or another word denoting darkness. In these verses
always takes the second position in the parallel structure which
indicates that covers much the same meaning only with a slightly
different nuance. Such observations have driven most scholars into
accepting the ṣlm-II reading, while others have tried to unify the meaning
of ‘darkness’ and the original reading of . This has happened by
interpreting as a superlative using as an intensifying element
(Thomas 1962). Unfortunately this endeavour has been unconvincing since
a grammaticalized superlative does not have the same potential for
darkening the positive meaning of as that a grammatically independent
and semantically fully functioning might have. Thus the clearly
negative connotation of cannot be explained (cf. note 13).
Nonetheless, I think that it is indeed possible to demonstrate that a -
etymology and a predominantly “darkness” meaning are not mutually
‘Darkness’28 would indeed fit better throughout all the occurrences of
. Cohen showed that is often paralleled or opposed to a word
meaning darkness (1996:289-291).29 This is indeed the case in 13 out of the
18 occurrences of . Secondly Cohen extrapolates this meaning to the
5 other cases, where such a parallel structure is not found. However, he is
too quick to assume that a meaning resulting from various similar parallel
28 Cohen’s meaning ‘darkness’ was further supported by Rendsburg (2001:189).
He adds that in the Egyptian Book of the Dead an expression similar to “valley
of darkness” ( ; Ps 23:4) is used: ı
͗nt kkt. This use of ‘darkness’ in a very
similar expression to Ps 23:4 is further indication that ‘darkness’ is appropriate.
29 Yet three other parallels, with (Job 38:17), (Ps 44:20) and
(Jer 2:6) remain unmentioned by Cohen in this regard.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 109
structures is the meaning. What Cohen has found here is not the meaning,
but one important aspect of the conceptual structure of .
Stone (2006) arrives at more or less the same observations.30 He listed
all the verses where occurs and grouped them according to the
acceptability of “darkness” as a translation.31 He shows that there are no
passages where “darkness” is impossible as a translation, only that it
sometimes appears to be less obvious than in other places. It should be
noted however that Stone’s primary approach consists in translation.
Consequently, he does not study the conceptual structure of , nor is
he looking for different semantic aspects of the word. However his findings
prove that ‘darkness’ is indeed an important aspect of . Especially
interesting, however, are his principles for interpretation:
Not only can death refer to darkness, but darkness can refer to death.
Words may have more than one meaning.
When two words are used in parallel, they are likely to be
Parallelism in poetry or in prose indicates emotion (Stone 2006:54-
These principles indicate that while “darkness” might be a good translation
(ranging from the required translation to a possible translation), it is by no
means the meaning of . Most scholars have tried to propose a one-
fits-all meaning for , but it seems that there are other aspects to take
into account also. Hence the conclusion of Stone: “I conclude that şlmwt
can be understood as ‘darkness’ in almost all, if not in all, cases. On the
other hand, a meaning related to death may apply in a few cases. To say
more would require a more detailed study of some of its occurrences”
(2006:57). In the following paragraph I will take up Stone’s challenge.
4.2 Other Aspects of
Thus it is obvious that ‘darkness’ should be an important aspect in the
mapping of the cognitive structure of . In order to amend this with
other aspects I have first conducted a collocation analysis. After
disregarding all particles and conjunctions and the collocates which occur
30 Stone (2006:53-57) refers nowhere to the article of Cohen, so I assume that his
article should be read independently of Cohen’s work.
31 These groups are “Passages in which ‘Darkness’ is Required”, “Passages in
which ‘Darkness’ is Very Likely” and “Passages in which ‘Darkness’ is
110 DAVID VAN ACKER
only once within a range of 5 words left and right of (within the same
verse), the following remained:
Word Frequency Verses
5 (27,8%) Job 10:21; 10:22; Isa 9:1; Jer 2:6
5 (27,8%) Job 3:5; 10:21; 34:22; Ps 107:10, 14
4 (22,2%) Job 12:22; Job 24:17; Isa 9:1; Jer 13:16
2 (11,1%) Job 10:22; 28:3
2 (11,1%) Job 24:17; Amos 5:8
2 (11,1%) Job 12:22; Ps 107:14
2 (11,1%) Ps 107:10; Isa 9:1
Table 1: Collocation Analysis of
From this analysis it is already obvious that besides words connected with
darkness and light ( , , and ), words related to places are the
other collocates ( , and the verbs and ).
To elaborate on this I have studied each individual verse qualitatively
and looked not only at the obvious synonyms and antonyms to (as
Cohen did), but also at the semantic subject of (i.e. what is described
as ) and concepts associated32 with in the context of the verse.
I first observed all these aspects for each verse (see Appendix 1 and 2).
Then I ordered all the concepts and noted all the verses where they occur
(see Appendix 3). Eventually I bundled these individual concepts into
abstract themes or conceptual categories33 which constitute the conceptual
32 This can be a word which is complemented by (e.g. May gloom and utter
darkness claim it [= the day] once more; here “the day” is something which
could be ); but it can also be an association due to adjacency (e.g. My face
is red with weeping, dark shadows ring my eyes; is parallel with
[red] and modifies [my eyelids] which is parallel to [my face].
However, it is also associated with weeping, implying that as the redness of the
face is invoked by weeping, so is the on the eyes).
33 One might wonder why these conceptual categories bear English labels. This
seems inconsistent, because these English words have their own meaning and
association which is not the same as their Hebrew counterparts. However, when
using Hebrew words, one might get the impression that the number of verses for
each label feature that specific Hebrew word – which might not be the case.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 111
frame of . The result of this categorisation is shown below (each
number indicates the number of individual instances of relating to
Parallel Antonym Subject Association TOTAL
Darkness 10 7 035 1 17
Location 3 0 9 5 17
Terror 4 0 3 5 12
Cover 0 1 0 1 2
Misc. 0 0 1 0 1
Table 2: The Semantic Themes related to
Obviously the categories in this table are semantically quite different from
each other: ‘darkness’ can be considered a meaning, while ‘location’ is a
semantic reference and ‘terror’ is a connotation of . However, I do not
want to compare these categories to each other, as I consider them to
complement each other eventually collecting all the necessary aspects of
the conceptual structure of . This structure thus consists of a meaning
of darkness framed by ‘location of terror’. These aspects can in every case
of be more or less present, depending on the context.
4.3 The ‘Terror’ and ‘Location’ Aspects
According to Niehr (1977:397-398) there are three contexts in which
occurs: 1. death and the underworld, 2. imprisonment and 3. theological
action. The first category contains most of the occurrences which I
classified as ‘terror’ and ‘location’. A few times co-occurs with
,36 this “land of (utter) darkness” is generally considered to refer to the
Therefore I have chosen to apply broad semantic labels in English to the
34 So if a verse has both ‘day’ and ‘light’ as opposites for the same instance of
, I originally noted this verse twice (cf. Appendix 3), but in the final table
presented here and ordered by theme ‘(opposite to) darkness’ I have counted this
verse only once.
35 The verses where a concept opposite to darkness is given (e.g. morning) are not
included here, but under ‘antonym’, however often this can also be considered
the subject of (e.g. “midnight [ ] is their morning”).
36 Job 10:21, 22; Isa 9:1; Jer 2:6.
112 DAVID VAN ACKER
underworld.37 Similarly, the parallelism in Job 38:17 between and
recalls the underworld. The same association can be made for the
“valley of darkness” in Ps 23:4. In this regard Tromp mentions as
part of Sheol (1969:140-143), but he also mentions “two mountains
bordering the nether-world” (1969:144). The “valley of darkness” could
easily refer to the valley between these mountains.38 Thus Ps 23:4 could be
a reference to the entrance to the realm of the death. This argument does
not seem unreasonable since it is very clear that the same concept ‘entrance
to the underworld’ is also present in the expression “gates of deepest
darkness” in Job 38:17. Two more instances must be added to these
references to the underworld, these mention as something deep or far
removed (Job 12:22 and Job 28:3, the latter refers to the depths and
darkness of an underground mine). The underworld is undoubtedly seen as
something terrible or inhospitable; that is clear from the above contexts. It
can therefore be said of that it refers to the terrible and inhospitable
nature of the underworld. In some of these instances, the terror is clearly
linked to a location (e.g. the wilderness in Jer 2:6 or the place of jackals in
Ps 44:20), but in some cases the terror itself takes the upper hand and the
location aspect is not present anymore (e.g. the terrors of darkness in Job
27:14 or the darkness on the eyes during weeping in Job 16:16). Anyway it
is clear that this use of with the implication of terror is not
unimportant. The terror is life-threatening. Niehr (1977:398) puts it as
follows “The term ṣalmāweṯ conjures up the notion of death caused by the
thirst and hunger to which the wanderer is subject”.39
The other two categories defined by Niehr as contexts of are
imprisonment and theological action. Both contexts are distinct derivatives
from the first context of death and underworld. The imprisonment
“resembles life in the underworld” (Niehr 1977:398). The theological
37 This association is even clearer in Akkadian, where the word for the underworld
is actually the same as the earth: erṣetu. Moreover, in the Akkadian literature the
underworld is often described as ‘a place of no return’ (erṣetu lā târi) (Katz
2014:343) a description which also occurs in one of the verses where is used
together with , Job 10:12: “before
I go to the place of no return, to the land of gloom and utter darkness” (NIV).
38 See also Rendsburg’s association of this passage with the Egyptian Book of the
Dead in footnote 28.
39 Unfortunately, even though Niehr (1977) acknowledges these aspects, he still
rejects the - reading. Yet it should be clear that exactly this interpretation
enables one to explain the notion mentioned by Niehr.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 113
language includes the passages which I labelled with the general theme of
darkness. It includes passages like Amos 5:8 where God is told to be the
one who can “turn darkness into morning” and draws on the antithesis
between life and death, light and dark, heaven and underworld.
4.4 - and ‘Darkness’
Having identified ‘terror (of death)’ and ‘location (of the underworld)’ as
additional aspects to the already established ‘darkness’-aspect of it
becomes clear how can be read as ‘darkness’ without having to
bypass the Masoretic vocalization. The reading on the other hand –
besides being difficult to explain etymologically – offers no explanation for
these observed semantic nuances.
It has been mentioned in the first overview of all the arguments that
compound nouns do not often occur in Hebrew, except for place names.
being a compound noun can thus be explained by interpreting it as a
place name or at least as having an aspect of locality to it. This locality
aspect, then, can be understood if would at one time have referred to
a (mythological) place. To support this I want to add possible evidence from
the Ugaritic texts. As I described above, ẓlmt in the parallel texts from the
Ba‘al cycle (KTU 1.8:7-9 and 1.4.7:54-56) should be identified with
in the Hebrew Bible. In these texts ẓlmt might either be a proper name or a
location. Both were probably related to Mot and to darkness. However, in
the Ugaritic text this word stands on its own, making its interpretation
difficult. In Hebrew on the other hand, is used more abstractly
modifying another noun or requiring a context describing it as a place. This
indicates that ẓlmt in the Ugaritic texts could still have a physical or
mythical referent and refer less to an abstract ‘darkness’. This can be
explained by what Barr calls ‘demythologizing’ (1974b:54). This change is
equal to the change of generalization described by Riemer (2010:374).
What once had a specific mythological referent now refers to a whole group
of referents. The mechanics of this change can also be explained by words
of the same author. We might say that the distant ‘realm of the death’ has
been internalised to a more close-by ‘darkness’ and that it has been
subjectified to fit with the user’s beliefs or own findings (Riemer 2010:380).
However, Barr proposes to subsequently read the demythologized or
generalized as ‘shadow of death’. But here I disagree with him. His
reading does not follow from the contexts in which is used. Moreover
he reads a compound as the sum of its components, which is not necessarily
114 DAVID VAN ACKER
the correct reading.40 I would therefore propose to read a demythologized
in the first place as ‘darkness’. Thus proposing that in the process of
generalisation the explicit meaning (underworld) and the implied meaning
(darkness) have switched places, resulting in the surface meaning of
‘darkness’ with the additional aspects of ‘terror (of death)’ and ‘location’
The following conjecture is a possible route from ẓlmt as a mythological
name ‘shelter/vicinity of Mot’ to ‘darkness’ with all its nuances.
a. Originally was a mythological name referring to one specific
b. At some point in the history of the meaning has started to
widen (generalise) but without any alteration to the word and
its reading - . This process has gone from ‘shelter of Mot’ to
‘cover of death’, reading as ‘the shelter of the covering by
shadow’. It is unclear which of the two following processes has taken
place. (1) Either this change was a semantic shift triggered when the
word came in use in Hebrew after being borrowed from another
language like Ugaritic (which would explain the compound nature of
the noun). (2) Or
has been in use as a place name in Hebrew
all along with the meaning generalising through time (given the
many examples of compound place names, this too would explain
the compound nature of the word). In either way it is clear that the
Ugaritic mythological name ẓlmt developed into the Biblical Hebrew
noun through a process of demythologization or
c. Eventually the meaning ‘darkness’ which is already implied in
‘shelter of Mot’ or underworld has gained the upper hand.
I would argue that it is possible that the meaning ‘darkness’ has emerged
from - . This argument, especially step b. might be further supported
by the fact that is related to ‘covering’ and ‘hiding’ in two verses.41
40 For some examples of this principle see Riemer (2010:57).
41 Job 34:22: No evildoers can hide in the , and Ps 44:20: you covered us with
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 115
It is impossible to fully grasp the whole meaning of and put it in a
one-size-fits-all translation.42 ‘Shadow of death’ is unsatisfying in many
ways. The translation ‘(deep) darkness’, is more appropriate in most
contexts, but it does not do justice to the rich conceptual structure of
and its etymology. Given the nuances identified in this study, it is advisable
to translate depending on the context with either ‘terrifying
darkness’, ‘deadly darkness’ or ‘(utter) dark place’.
1. Analysis of the Verses
The translations are taken from the NIV translation of the Bible. Italics is
the translation of ; dotted underlined are the parallels or antonyms of
; underlined are associated concepts of ; square brackets are
implications which are not specifically named in the text.
42 Among the different translations “shadow of death” is most present (KJV, ESV,
SVV, LSG, ELB) along with “(deepest, blackest or utter) darkness” (NIV, ESV,
CEB, NBV, NBG, ELB), “(deep or dark) shadow” (NIV, ESV, CEB), “(dark or
deep) gloom” (ESV, CEB) and “pitch-dark” (CEB, NBG). Some (KJV, SVV,
NBV, LSG) translate almost consistently with some kind of “shadow of death”,
while other translations provide many different translations depending on the
context, but mostly revolving around the idea of darkness (CEB, NBV, ELB).
However, this does not necessarily imply that the hypothesis underlies this
translation. For the translations using ‘darkness’ also use translations where
‘shadow’ is evident (e.g. “deep darkness” in Job 3:5 [ESV], while “deep shadow”
in Job 10:21 [ESV]; or “Finsternis” in Job 3:5 [ELB], while “Todesschatten” in
Job 10:21 [ELB]). Moreover, in some translations other aspects of the word
become evident. These translations include “deep gloom” (ESV, CEB, NBV),
“pitch dark” (CEB, NBV) or “dark holes” (NBV). Another aspect of which
is more present in translations, is the “utter”, “deep” or “blackest” epithet
distinguishing it from other semantically related words like or (usually
translated darkness) and (thick darkness). Indirectly some translations also
provide a location aspect (e.g. “Donkere krochten” in Ps 107:10 [NBV] or “<das
Land> der Finsternis” in Job 10:22 [ELB]), the word “death” on the other hand
is rarely used outside of “shadow of death”.
116 DAVID VAN ACKER
May gloom and utter darkness claim it [= the day] once more;
may a cloud settle over it; may blackness overwhelm it.
before I go to the place of no return, to the land of gloom and
to the land of deepest night, of utter darkness and disorder, where
even the light is like darkness.
He reveals the deep things of darkness and brings utter darkness
into the light.
My face is red with weeping, dark shadows ring my eyes;
(16: In the dark, thieves break into houses, but by day they shut
themselves in; they want nothing to do with the light.) For all of
them, midnight is their morning; they make friends with the
terrors of darkness.
Mortals put an end to the darkness; they search out the farthest
recesses for ore in the blackest darkness [Place].
There is no deep shadow, no utter darkness, where evildoers can
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 117
Have the gates of death been shown to you? Have you seen the
gates of the deepest darkness?
Even though I walk through the darkest valley, I will fear no evil,
for you are with me; your rod and your staff, they comfort me.
But you crushed us and made us a haunt for [place of] jackals;
you covered us over with deep darkness.
Some sat in darkness, in utter darkness, prisoners suffering in
He brought them out of [Place] darkness, the utter darkness, and
broke away their chains.
The people walking in darkness have seen a great light; on those
living in the land of deep darkness a light has dawned.
They did not ask, ‘Where is the LORD, who brought us up out of
Egypt and led us through the barren wilderness, through a land of
deserts and ravines, a land of drought and utter darkness, a land
where no one travels and no one lives?’
118 DAVID VAN ACKER
Give glory to the LORD your God before he brings the darkness,
before your feet stumble on the darkening hills. You hope for
light, but he will turn it to utter darkness and change it to deep
He who made the Pleiades and Orion, who turns midnight into
dawn and darkens day into night, who calls for the waters of the
sea and pours them out over the face of the land – the LORD is his
2. The Semantic Observations
To what is
Job 3:5 Day Cloud;
Job 10:22 ;
Job 12:22 Light Deep
Job 16:16 Redness Eyes Weeping
Job 24:17a Day; Light;
Job 24:17b Terrors
Job 28:3 ;
Place Ore; Far
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 119
Job 34:22 Hide Place Evildoers
Job 38:17 Death Gates
Ps 23:4 Valley Evil
Ps 44:20 Place;
Ps 107:10 Iron chains Prisoners
Ps 107:14 Place Chains
Isa 9:1 Light Land
Jer 2:6 Barren
Jer 13:16 ; Light
Amos 5:8 ;night Dawn; Day
3. The Semantic Observation per Concept
When is it
120 DAVID VAN ACKER
Night Amos 5:8
Color Job 16:16
Cloud Job 3:5
Valley Ps 23:4
Ps 44:20 Job 28:3;
Prison Ps 107:10
Prisoner Ps 107:10
Gates Job 38:17
Weeping Job 16:16
Crushing Ps 44:20
Disorder Job 10:22
Terror Jer 2:6 Job
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 121
Death Job 38:17
Jackals Ps 44:20
Drought Jer 2:6
Cover Covering Ps 44:20
Hide Job 34:22
Misc. Eyes Job 16:16
Balogh, C 2014. Historicising Interpolations in the Isaiah-Memoir. VT 64/4, 519-538.
Barr, J 1974a. Etymology and the Old Testament, in: Barr, J, Beuken W A M, Gelston,
A, et al. (eds). Language and Meaning: Studies in Hebrew Language and Biblical
Exegesis (Oudtestamentische studiën). Leiden: Brill, 1-28.
Barr, J 1974b. Philology and Exegesis, Some General Remarks, with Illustrations from
Job, in: Brekelmans, C, Barr, J & De Boer, P A H (eds). Questions disputées
d’Ancien Testament: méthode et théologie. Gembloux: Duculot, 39-61.
Bordreuil, P & Caquot, A 1980. Les textes en cunéiformes alphabétiques découverts en
1978 a Ibn Hani. Syria 57/2, 343-373.
CAD = Gelb, I J, Landsberger, B & Oppenheim, A L (eds) 1956-2010. The Assyrian
Dictionary of the Oriental Institute of the University of Chicago. 21 Vols. Chicago:
The Oriental Institute.
CEB = 2011. Common English Bible. Nashville: Common English Bible.
Clines, D J A 1974. The Etymology of Hebrew ṣelem. JNSL 3, 19-25.
Clines, D J A 1989. Job 1-20 (Word Bible Comentary 17). Dallas: Word Books.
Clines, D J A (ed.) 1993. The Dictionary of Classical Hebrew. Sheffield: Sheffield
Cohen, C 1996. The meaning of “Darkness”: A study in Philological Method, in:
Fox M V (ed.). Texts, Temples, and Traditions: A Tribute to Menahem Haran.
Winona Lake: Eisenbrauns, 287-309.
Del Olmo Lete, G & Sanmartín, J 2003. A Dictionary of the Ugaritic Language in the
Alphabetic Tradition. 2nd revised ed. Leiden: Brill.
Dietrich, M & Loretz, O 1983. Neue Studien zu den Ritualtexten aus Ugarit. Ugarit-
Forschungen: Internationales Jahrbuch für die Altertumskunde Syrien-Palästinas
Driver, S R & Gray, G B 1964. A Critical and Exegetical Commentary on the Book of
Job together with a New Translation. Reprint 1950. Edinburgh: Clark.
122 DAVID VAN ACKER
ELB = 2006. Elberfelder Bibel. Witten: SCM R. Brockhaus.
Eybers, I H 1972. The Root ṣ-l in Hebrew Words. JNSL 2, 23-36.
ESV = 2016. The Holy Bible: English Standard Version. Wheaton: Standard Bible
Hankins, D 2013. Wisdom as an Immanent Event in Job 28, Not a Transcendent Ideal.
VT 63/2, 210-235.
Haupt, P 1914. Die Schlacht von Taanach. BZAW 27, 192-225.
Jones, S C 2009. Rumors of Wisdom: Job 28 as Poetry. Berlin: Walter de Gruyter.
Katz, D 2014. Unterwelt A. I., in: Ebeling, E & Meissner, B (eds). Reallexikon der
Assyriologie und Vorderasiatischen Archäologie. Vol. 14. Berlin: Walter de
KJV = 2009. The Holy Bible: King James Version. Electronic Edition of the 1900
Authorized Version. Bellingham, WA: Logos Research Systems, Inc.
LSG = 1988. La Sainte Bible: Louis Segond Version. Online Bible Foundation and
Woodside Fellowship of Ontario.
Mazzini, G 1999. On the Meaning of Salmawet: A Discussed Word in the Old
Testament. Studi epigrafici e linguistici sul Vicino Oriente antico 16, 79-83.
Michel, W L 1984. Şlmwt, “Deep Darkness” or “Shadow of Death”. Biblical Research
Mittmann, S 1980. Aufbau und Einheit des Danklieds Psalm 23: Friedrich Lang in
herzlicher Verehrung zum 65. Geburtstag. ZTK 77/1, 1-23.
NBG = 1951. De Bijbel: Vertaling 1951. Nederlands Bijbelgenootschap.
NBV = 2004. De Nieuwe Bijbelvertaling. Nederlands Bijbelgenootschap.
Niehr, H 1977. , in: Botterweck, G J & Ringgren, H (eds). Theological Dictionary
of the Old Testament. Vol. 12. Grand Rapids: Eerdmans, 396-399.
NIV = 2011. The New International Version. Grand Rapids, MI: Zondervan.
Pardee, D 2009. A New Join of Fragments of the Baal Cycle, in: Schloen, J D (ed.).
Exploring the Longue Duree: Essays in Honor of Lawrence E. Stager. Winona
Lake: Eisenbrauns, 377-390.
Price, J D 1997. , in: Jenni, E & Westermann, C (eds). Theological Lexicon of the
Old Testament. Vol. 2. Peabody: Hendrickson, 807-810.
Rendsburg, G A 2001. Hebrew Philological Notes (II). HS 42, 187-195.
, AN ETYMOLOGICAL AND SEMANTIC RECONSIDERATION 123
Rendsburg, G A 2013. Phonology: Biblical Hebrew, in: Khan, G (ed.). Encyclopedia of
Hebrew Language and Linguistics. Vol. 3. Leiden: Brill, 100-109.
Riemer, N 2010. Introducing Semantics. Cambridge: Cambridge University Press.
Smith, M S & Pitard, W T 2009. The Ugaritic Baal Cycle (VTS 114). Leiden: Brill.
Stone, A P 2006. Does “Shadow of Death” mean “Deep Darkness”? Biblical Research
SVV = 1997. Staten Vertaling. Electronic ed. Bellingham, WA: Logos Bible Software.
Thomas, D W 1962. in the Old Testament. JSS 7/2, 191-200.
Tromp, N J 1969. Primitive Conceptions of Death and the Nether World in the Old
Testament. Rome: Pontifical Biblical Institute.
Whitley, J B 2015. in Amos 4:13: New Evidence for the Yahwistic Incorporation
of Ancient Near Eastern Solar Imagery. JBL 134/1, 127-138.
Yon, M 1997. La Cité d’Ougarit Sur Le Tell de Ras Shamra. Paris: Recherche sur les