The Semantics of Biblical Language Redux

Cameron Boyd-Taylor

1. Introduction
It was in 1961 that James Barr published The Semantics of Biblical Language, in
which he laid the foundations for a linguistically oriented approach to biblical
lexicography. 1 Most biblical scholars have encountered this seminal work in one
way or another and have no doubt profited accordingly, so there is no need for
me to rehearse the main lines of Barr’s argument. My intention in this paper is to
revisit a number of key points and relate them to the semantics of the Greek Bible.
The burden of my argument is that the field of Septuagint studies has yet to
work out the implications of the biblical semantics movement pioneered by Barr.
I attribute this lag to the fact that Barr’s criticisms were leveled primarily at bibli-
cal theology as it was practiced at the time;2 it is only very recently that much
sustained work has been done on the so-called theology of the Septuagint. Hence,
some of the problems Barr was addressing are only now arising. At the same time,
the task for Septuagint studies must be distinguished from that of biblical theology
proper. What is required, I suggest, is an approach to the Septuagint that squarely
addresses the semiotic opposition between translational and nontranslational
literature,3 an opposition that I take to be axiomatic for the field.

1. James Barr, The Semantics of Biblical Language (Oxford: Oxford University Press, 1961).
2. See James Barr, The Concept of Biblical Theology: An Old Testament Perspective (Min-
neapolis: Fortress, 1999), 232: “As is clear, the faults in the use of linguistic evidence which I
detected and criticized were for the most part located in ‘biblical theology’ as it was practised at
that time, and I said so.”
3. See Gideon Toury, “The Meaning of Translation-Specific Lexical Items and Its Repre-
sentation in the Dictionary,” in Translation and Lexicography: Papers Read at the EURALEX Col-
loquium Held at Innsbruck 2–5 July 1987 (ed. Mary Snell-Hornby and Ester Pöhl; Amsterdam and
Philadelphia: Benjamins, 1989), 45–53, here 45; see also Gideon Toury, Descriptive Translation
Studies and Beyond (Benjamins Translation Library 4; Amsterdam and Philadelphia: Benjamins,
1995), 274–79.

41

HiebertA.indd 41 8/30/2010 7:52:12 PM

 42 “Translation Is Required”

This, for me, was the signal lesson of the NETS project: the realization that the
critical study of a translation, any translation, raises hermeneutic issues of a differ-
ent sort from those occasioned by an original composition. To the extent that we
are able to speak meaningfully of a theology of the Septuagint, if indeed we can,
the manner in which we proceed will be very different from our discussion of the
Hebrew Bible. I shall take up this theme presently. For now I return to Barr, who,
more than anyone else, has shown the way forward. To focus the discussion, I shall
consider the semantic field marked out by the English word “hope,” with specific
reference to the use of the verb ἐλπίζω in the Greek Psalter.

2. Word and Concept

A central tenet of Barr’s approach is the distinction between word and concept.4
While a word may, in a given context, express a specific concept, it does not follow
that its semantic field reflects the structure of that concept.5 Of course, in theol-
ogy we frequently use words as a shorthand for key biblical themes. We may, for
instance, speak of a theology of hope. Here the word “hope” does double duty as
a label for a complex discursive structure, though this is not in itself problematic.
Barr’s point is that if we carry this practice over into the analysis of biblical lan-
guage, we risk all manner of conceptual confusion.
The problem is compounded when we turn to the language of the Septuagint.
For here there is a temptation to assume that through the process of Hebrew–
Greek translation concepts expressed in the source language somehow passed over
into lexical meanings within the target language. A celebrated instance of this phe-
nomenon is the Greek verb ἐλπίζω. Ceslas Spicq, for instance, claims that “a veri-
table semantic revolution is effected by the LXX,” which gives the word “a strictly
religious meaning.”6 In Attic Greek, ἐλπίζω carries the near universal meaning
of expectation in relation to the future—the expectation in question may be good
or bad as the case may be.7 But in the Septuagint, particularly the Greek Psalter, a
very different attitude is evident. As Spicq writes, “Hope, which is always directed

4. David Alan Black, Linguistics for Students of New Testament Greek: A Survey of Basic
Concepts and Applications (Grand Rapids: Baker, 1995), 123.
5. James Barr, Biblical Words for Time (SBT 33; London: SCM, 1962), 110: “In other words
we once again repudiate the treatment of a word as a ‘concept’ in the way which has been normal
in modern biblical theology. Such a method normally leads naturally to an attempt to regard the
location of the various items on the vocabulary grid of a particular language as reproducing the
essential elements of a thought structure in the typical mind of the members of the speech group
concerned.”
6. Ceslas Spicq, Theological Lexicon of the New Testament (trans. and ed. James D. Ernest;
3 vols.; Peabody, Mass.: Hendrickson, 1994), 1:485, s.v. ἐλπίζω.
7. Walther Zimmerli, Man and His Hope in the Old Testament (SBT 2/20; London: SCM,
1971), 2.

HiebertA.indd 42 8/30/2010 7:52:13 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 43

toward God, is no longer any expectation whatsoever, but a sure and certain con-
fidence in YHWH.”8
In the background of Spicq’s remarks lies the influence of Gerhard Kittel’s
Theological Dictionary of the New Testament (TDNT), the prime target of Barr’s
critique. In his contribution to the entry for ἐλπίζω, Rudolf Bultmann had writ-
ten that, in the Septuagint, “The righteous are always referred to what God will do,
so that hope is not directed to anything specific, nor does it project its own view
of the future, but it consists rather in general confidence in God’s protection and
help.”9 This attitude is to be distinguished from the Greek habit of associating hope
with uncertainty, fantasy and folly.10
How did this supposed semantic change come about? The underlying assump-
tion of the TDNT article is voiced by Nigel Turner: ἐλπίζω derived its biblical
meaning from x+b.11 The Hebrew verb is often found in prayer and song formu-
lae, where it describes security in God. In the Psalter it regularly occurs in contexts
which assume that “in times of distress there is no way for man to survive but to
take refuge in Yahweh, to trust in him, and to have confidence in him.”12 Since x+b
is typically rendered by ἐλπίζω in the Greek Psalter, Turner implies, the latter
took up the theological sense carried by its Hebrew counterpart.
This, then, is the locus of the semantic revolution to which Spicq refers. The
word ἐλπίζω has shed the garments of Greek pessimism and put on the splendid
raiment of a theology of hope. Hope, writes Spicq, is now “a matter of finding one’s
refuge in YHWH; of having full and complete confidence in him.”13 The principal
heir of this language, this lexicon, and this faith, Spicq continues, is the Apostle
Paul, for whom hope pertains to “the whole economy of the new covenant.”14 In this
respect, the object of hope is now decidedly eschatological. Bultmann expresses
this idea with characteristic force: “Christian hope rests on the act of salvation, and
since this is eschatological, hope is itself an eschatological blessing.”15
And so it is. But, as James Barr taught us, when we speak this way we are talk-
ing about a concept, not a word. There can be no doubt that the Pauline corpus is
characterized by a theology of hope.16 That its distinct verbal texture is indebted,
at least in part, to the language of the Septuagint is also not in question. What is

8. Spicq, Theological Lexicon, 1:485.

9. Rudolf Bultmann, “ἐλπίς, ἐλπίζω,” TDNT 2:517–33, here 522–23.
10. In classical Greek literature, it is a truism that the hopes of mortals are uncertain and
misleading: in a word, hope is blind (Bultmann, “ἐλπίς,” 517–19). See also Zimmerli, Man and
His Hope, 2–3; and Michel Despland, The Education of Desire: Plato and the Philosophy of Religion
(Toronto: University of Toronto Press, 1985), 281.
11. Nigel Turner, Christian Words (Nashville: Thomas Nelson, 1981), 214.
12. Alfred Jepsen, “x+b,” TDOT 2:88–94, here 91.
13. Spicq, Theological Lexicon, 1:485.
14. Ibid., 489.
15. Bultmann, “ἐλπίς,” 532.
16. For Paul, hope is based on the gospel and centered in God; it is a present endowment of

HiebertA.indd 43 8/30/2010 7:52:13 PM

 44 “Translation Is Required”

questionable is the idea that underlying this language is a fundamental change in

the semantics of the word ἐλπίζω. Yet this assumption has made its way uncon-
tested into both the lexicography and the exegesis of the Greek Bible. It needs to
be closely examined.

3. Semantics in Biblical Research

If semantic analysis is to escape the pitfalls identified by Barr, a number of prin-
ciples need to be observed. In his monograph on semantics in biblical research,
John Sawyer identifies four.17
1. An adequate definition of context must precede every semantic statement.18
As Sawyer indicates, contextualization is a key factor in semantic description. This
poses something of a challenge to biblical scholars, since Scripture, whether taken
as a whole or in separate units, may be contextualized in so many different situ-
ations.19 The first step in any semantic study is thus to make clear which situation
has been chosen.20 For a lexical study of the Greek Psalter, I would suggest that the
context of its production as a translation commends itself for two reasons: first,
most exegetes are interested in precisely this situation; and, second, it is a context
over which some degree of methodological control may be exercised. It should be
noted, however, that other situations are also defensible.
2. Semantic statements must be primarily synchronic.
This is to say that diachronic statements regarding semantic change can be made
only after an adequate synchronic description has been completed.21 How this
is undertaken will depend in part on the context one is describing. Since I have
selected the translation of the Greek Psalter as the situation under description in
this study, my semantic analysis must take into account the fact that two sets of
linguistic norms are at work: those proper to the lexicon of Hellenistic Greek as
such, and those underlying the process of Hebrew–Greek translation that gave rise
to the text.22

the faithful, something not shared by the unbeliever; in hope we believe despite appearances and
look ahead to the life beyond; in hope we rejoice (Turner, Christian Words, 214–15).
17. John F. A. Sawyer, Semantics in Biblical Research: New Methods of Defining Hebrew
Words for Salvation (SBT 2/24; London: SCM, 1972).
18. Ibid., 112.
19. Ibid., 113.
20. Ibid., 114.
21. Ibid.
22. See Cameron Boyd-Taylor “Toward the Analysis of Translational Norms: A Sighting
Shot,” BIOSCS 39 (2006): 27–46.

HiebertA.indd 44 8/30/2010 7:52:13 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 45

3. Semantic universals operate in biblical languages as much as in any other

language.
Although Sawyer deals exclusively with Hebrew, his third principle applies with
special force to biblical Greek—it is not to be treated as a special case. This raises
a fundamental point. Since the Greek Psalter is a translation, Sawyer’s third prin-
ciple directs us to view its language in relation to the universals described by trans-
lation theorists. One such universal is the phenomenon of interference, which is to
say, a signal feature of all translation literature is linguistic interference from the
source language.23 Forms and structures occur that are seldom if ever encountered
in original compositions. The Greek Psalter, like any other translation, cannot be
taken as a straightforward sample of the target language.
4. A structural approach is required.
When Sawyer was writing, structuralism was in vogue. We now find ourselves in
a post-structuralist world, but the basic point stands. Semantic change should be
understood in terms of shifting forms and relationships.24 To capture this, we want
an appropriate semantic model. Like Sawyer, I take the proper focus of a lexical
study to be intralingual, that is, it should describe the relations between words.25
To my mind Sawyer’s four principles neatly capture the core implications of
the biblical semantics movement. Using them as guidelines, I shall now consider
the semantics of ἐλπίζω. What follows constitutes not a comprehensive analysis
but a sketch propaedeutic to such a study.

4. The Semantics of Hope in the Greek Psalter

In the field of biblical semantics, most are agreed that the pioneering work of
Johannes P. Louw and Eugene A. Nida represents a major step forward. Their
Greek-English lexicon classfies the vocabulary of the New Testament according
to semantic domains,26 and although the analysis has a number of limitations, it

23. See Cameron Boyd-Taylor, “Calque-culations: Loanwords and the Lexicon,” BIOSCS
38 (2005): 79–99, here 83. Cf. Toury, Descriptive Translation Studies, 275.
24. It is misleading to treat a word on its own, as semantic shifts are typically the result
of complex processes disseminating across whole networks of words. Tom McArthur, ed., The
Oxford Companion to the English Language (Oxford: Oxford University Press, 1992), 913.
25. Sawyer, Semantics in Biblical Research, 115. On this view, it is the relations between
words that matter, relations such as incompatibility, antonymy, hyponymy, consequence, and
synonymy. The meaning of linguistic items is thus defined without reference to extralingual fea-
tures. This is not to deny that denotation is crucial for definitional purposes, but simply to say
that it must follow a description of sense relations. I do not, however, intend to be doctrinaire
about this and will address various aspects of word meaning as the need arises.
26. Johannes P. Louw and Eugene A. Nida, Greek-English Lexicon of the New Testament
Based on Semantic Domains (New York: United Bible Societies, 1989).

HiebertA.indd 45 8/30/2010 7:52:13 PM

 46 “Translation Is Required”

provides a useful point of departure for the present discussion, for it is an excellent
example of the sort of structural approach favored by Sawyer.
Louw and Nida distinguish three types of semantic features: shared, distinc-
tive, and supplementary. Shared features are those elements of the meaning of a
lexeme held in common with other lexemes; distinctive features separate mean-
ings one from another; and supplementary features are those that are relevant
only in certain contexts or play a primarily connotative or associative role. Ana-
lyzing lexemes according to these features gives rise to a taxonomy of the lexicon.
Lexemes are classified according to higher-order domains, such as plants, ani-
mals, and kinship terms. Each domain is then divided into sets of subdomains,
which are related to one another largely in terms of greater and lesser degrees of
specificity.
The verb ἐλπίζω appears twice in Louw and Nida’s taxonomy. In the first
instance, it occurs in domain 25, attitudes and emotions, subdomain D, hope,
look forward to. Two other New Testament verbs are found in subdomain D,
ἀπεκδέχομαι, in the sense, to await eagerly or expectantly for some future event;
and προελπίζω, to hope in a prior manner. Subdomain D clusters with that of
desire, love, willingness, and eagerness.
The second occurrence of ἐλπίζω is located in domain 30, to think, subdomain
C, to think concerning future contingencies. Also in this domain is προσδοκάω, in
the sense, to expect something to happen, whether good or bad, and ἐκδέχομαι,
in the sense, to expect something to happen, often implying waiting. Subdomain C
clusters with the following subdomains: to think about, to intend or purpose, and
to decide or conclude.
Louw and Nida’s analysis represents a very good start. Of course, it can—
indeed, it must—be improved upon.27 But a perusal of the standard Greek lexica
indicates that their classification of ἐλπίζω, although based on a small corpus, is
nevertheless consistent with the results of Greek lexicography to date. But if we
turn to what is undoubtedly the most linguistically sophisticated lexicon of the
Septuagint, that of Takamitsu Muraoka,28 we find two distinct senses of ἐλπίζω
that do not appear in the other lexica. For present purposes, I need address only
one of them, no. 1, “to put trust (in), to count (on)”.
This meaning locates ἐλπίζω in domain 31 of the Louw-Nida taxonomy, to
hold, believe, trust, subdomain I, to trust or rely. It groups ἐλπίζω with πείθω
(perfect stem only), in the sense to believe in something or someone to the extent of
placing reliance or trust in or on, and πιστεύω, in the sense, to believe to the extent
of complete trust and reliance.

27. First, the evidence of the New Testament should be analyzed against the background of
contemporary Greek usage. Second, a wider array of semantic relations needs to be considered.
28. Takamitsu Muraoka, A Greek-English Lexicon of the Septuagint: Chiefly of the Penta-
teuch and the Twelve Prophets (Leuven: Peeters, 2002).

HiebertA.indd 46 8/30/2010 7:52:13 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 47

Now this classification of ἐλπίζω is consistent with the theological diction-

aries of Bultmann and Spicq, whom Muraoka cites. But, as these scholars were
aware, it implies semantic change in the biblical lexicon. Not to put too fine a
point on it, Muraoka’s sense number 1 lacks both the prospective and affective
force identified by Louw-Nida; on his analysis, ἐλπίζω is no longer bound up
with future contingencies and our attitude toward them. To place trust in some-
one obviously has a future reference: my trust anticipates your good faith. But
the attitude of trust is primarily oriented to the continuation of present realities,
that is, one’s belief in the reliability, truth, ability, or strength of someone else. The
verb ἐλπίζω, on the other hand, picks out a state of expectation rather than one
of belief. It looks to an anticipated future, whether desired or dreaded, and may
be used in opposition to τὰ παρόντα, “present circumstances” (e.g., Philo, Leg.
3.86). Pseudo-Plato defines the noun ἐλπίς as προσδοκία ἀγαθοῦ, “expectation
of (a) good” (Def. 416a).29
Is there any evidence in the Greek Psalter for a shift away from this prospec-
tive sense? Muraoka gives eleven examples of contexts in which his suggested
meaning seems plausible enough.30

1. LXX Ps 32:20–21: Our soul waits for the Lord, because he is our helper and
protector, (21) because in him our heart will be glad, and in his holy name
we hoped/trusted [x+b].
2. LXX Ps 41:6: Why are you deeply grieved, O my soul, and why are you
throwing me into confusion? Hope/trust [lxy] in God, because I shall
acknowledge him; my God is deliverance of my face.
3. LXX Ps 117:9: It is better to hope/trust [hsx] in the Lord than to hope/
trust [hsx] in rulers.
4. LXX Ps 61:11: Put no hope/trust [x+b] in wrong, and do not long for what
is robbed; wealth, if it flows, do not aadd hearta [a–a possibly set your heart
on it].
5. LXX Ps 77:21–22: Therefore, the Lord heard and was put out, and a fire was
kindled in Iakob, and anger mounted against Israel, (22) because they had
no faith in God nor did they hope/trust [x+b] in his saving power.
6. LXX Ps 118:42: And I shall have a word for those who reproach me, because
I hoped/trusted [x+b] in your words.
7. LXX Ps 51:10: But I am like a fruitful olive tree in the house of God. I
hoped/trusted [x+b] in the mercy of God forever, even forever and ever.
8. LXX Ps 146:11: The Lord is pleased with those who fear him and with
those who hope/trust [lxy] in his mercy.

29. See Bultmann, “ἐλπίς,” 518.

30. In this list of readings, apart from alternative renderings that are signified by italics,
the English translation is that of NETS (A New English Translation of the Septuagint [ed. Albert
Pietersma and Benjamin G. Wright; Oxford and New York: Oxford University Press, 2007]).

HiebertA.indd 47 8/30/2010 7:52:13 PM

 48 “Translation Is Required”

9. LXX Ps 43:7: For not in my bow shall I hope/trust [x+b], and my sword
will not save me.
10. LXX Ps 7:2: O Lord my God, in you I hoped/trusted [hsx]; save me from
all my pursuers, and rescue me . . .
11. LXX Ps 12:6: But I hoped/trusted [x+b] in your mercy; my heart shall
rejoice in your deliverance. I will sing to the Lord, my benefactor, and
make music to the name of the Lord, the Most High.

If we substitute the word “trust” for “hope,” the resulting English gloss works well
enough in each context. But as Moisés Silva has argued, contextual fit is not an
altogether reliable semantic criterion; it is by no means sufficient for establish-
ing semantic change in a word.31 We must proceed to ask whether the contextual
meaning characterizes the word as such, and, conversely, whether the word as such
introduces this meaning into the context. To answer this question we must inves-
tigate the precise relationship between word and context. In the case of the Greek
Psalter, this relationship is bound up with the fact that the text is a translation.
Let us consider the examples given by Muraoka. In each instance the special
biblical meaning “to trust” that he has assigned to ἐλπίζω is, arguably, a reflex of
the underlying Hebrew context. Do we assume that in each case the Greek verb
carries the meaning of its Hebrew counterpart? Or has Muraoka read a Hebrew
meaning into a Greek word? To paraphrase James Barr: unless we know why the
translator made the choices he did, we can hardly draw semantic inferences.32 This
means taking into account translation technique.33
As it happens, the translator of the Psalter tends to replace Hebrew words with
Greek equivalents according to certain default lexical matches, much in the man-
ner of an interlinear translation.34 This means that in any given instance there may
be considerable semantic tension between a Greek word and its larger context, the
result of interference from the Hebrew. The verb ἐλπίζω is no exception.
It occurs 73 times in Rahlfs’s edition of the Greek Psalter and renders seven
different Hebrew verbs. Three of these matches underlie the texts identified by
Muraoka: x+b, “to trust,” occurs 46 times in the Psalter and is rendered by ἐλπίζω
37 times; hsx, “to seek refuge,” is rendered by ἐλπίζω 20 out of 25 times, and lxy,
“to wait or await,” is rendered by either ἐλπίζω or its cognate ἐπελπίζω 18 out of

31. Moisés Silva, Biblical Words and Their Meaning: An Introduction to Lexical Semantics
(rev. and exp. ed.; Grand Rapids: Zondervan, 1994), 199–200.
32. James Barr, “Common Sense and Biblical Language,” Bib 49 (1968): 377–87, here 379.
33. See Barr, Words for Time, 119: “We may however add a reminder that the language of
the LXX occupies a peculiar place; it is a translation register, which varied in important respects
from the actual speech of the Hellenistic Jews, and which at certain points and in certain respects
can be understood only in relation to the translation techniques being used in the various sec-
tions.”
34. Albert Pietersma, “Psalms: To the Reader,” in NETS, 542–47, here 542.

HiebertA.indd 48 8/30/2010 7:52:13 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 49

18 times. For each of these Hebrew verbs, therefore, ἐλπίζω serves as a preferred
rendering, a kind of default. Hence, in any given context one does not want to read
too much into its occurrence: quite often the Greek word has been selected by the
translator simply because its Hebrew counterpart is present.
There are exceptions, however, and these can prove telling. We may take the
translator’s rendering of x+b as a case in point. As Walther Zimmerli observed
in his Göttingen lectures, the use of ἐλπίζω as a default is somewhat surprising.35
All else being equal, we might have expected πείθω. The translator often follows
the lead of the Pentateuch,36 and in Deut 28:52 x+b is rendered by the perfect of
πείθω. Of course, on the assumption of semantic change, ἐλπίζω and πείθω have
become near synonyms, which would neatly account for the usage of the Psalter.
But does the translator actually use them as such? Interestingly enough, in the
seven instances where he decides against matching x+b with ἐλπίζω, he supplies
πείθω. They are not, however, free variants, for if we look at the contexts in which
πείθω is used instead of ἐλπίζω, a pattern emerges that tells against this idea.
In four out of the seven cases, the object of x+b is something people mistakenly
place their trust in or rely upon, whether wealth (49:6, LXX 48:7), idols (115:8,
LXX 113:16; 135:18, LXX 134:18) or mere mortals (146:3, LXX 145:3). Conversely,
when ἐλπίζω renders x+b, the object of hope is either God or it is unspecified
(with God being implied). The one exception proves the rule, for here ἐλπίζω and
πείθω are used in contrastive stichs (118:7–8, LXX 117:7–8). There is no indica-
tion, then, that ἐλπίζω and πείθω are interchangeable; in fact, their distributions
are quite distinct. Clearly we have no reason to posit a new meaning for ἐλπίζω
and must assume that it continues to carry its conventional sense.37
When we look at the other two matches underlying Muraoka’s list, we find
nothing to disconfirm this. As in the case of x+b, the translator’s rendering of hsx
is somewhat unexpected. In Deut 32:37 the Hebrew verb is rendered by the perfect
of πείθω, a match taken up by other translators—including, on four occasions,
the one responsible for the Psalter. Given that the Hebrew verb is used regularly
as a formula of trust, this is not a surprising equivalency. What is interesting is
that the Greek psalmist has recourse to πείθω so rarely. Judging by his rendering

35. Zimmerli, Man and His Hope, 9.

36. See Emanuel Tov, “The Impact of the Septuagint Translation of the Torah on the
Translation of the Other Books,” in Mélanges Dominique Barthélemy: Études bibliques offertes à
l’occasion de son 60e anniversaire (ed. Pierre Casetti et al.; OBO 38; Göttingen: Vandenhoeck &
Ruprecht, 1981), 577–92; Jan Joosten, “The Impact of the Septuagint Pentateuch on the Greek
Psalms,” in XIII Congress of the International Organization for Septuagint and Cognate Studies,
Ljubljana, 2007 (ed. Melvin K. H. Peters; SBLSCS 55; Atlanta: Society of Biblical Literature,
2008), 197–206.
37. It is interesting to note that while Zimmerli (Man and His Hope, 9) accepts this conclu-
sion, he argues that the Greek translator has stressed that aspect of hope which involves personal
surrender, that is, trust. Yet surely this is to get things backwards. What the psalmist has done is
stress that aspect of trust that involves expectation.

HiebertA.indd 49 8/30/2010 7:52:14 PM

 50 “Translation Is Required”

of hsx as σκεπάζω in 61:5 (LXX 60:5), where it is used in a metaphor of being

sheltered by God’s wings, the translator was not altogether deaf to the claims of
context. Yet, as we have seen, in the vast majority of contexts, where it represents
an expression of confidence in YHWH, he renders it by ἐλπίζω. Here again the
translator favors the equivalent that carries the prospective sense.
Finally, there is the match with lxy, for which Gen 4:26 may have provided
the model. Here the source item carries a prospective sense, “to wait,” so there is
semantic overlap with ἐλπίζω. This, incidentally, confirms that the Greek word
is carrying its conventional meaning. In every single context the translator sup-
plies either ἐλπίζω or ἐπελπίζω. The verb πείθω is never used as a match, even
though there are numerous contexts in which it might fit.
Obviously there is more to be said about the translator’s use of ἐλπίζω. Hav-
ing identified the default matches, one ought to consider all the exceptions and
examine the linguistic context of each in detail. Then one should examine the
translator’s handling of words in related subdomains. In a fine lexical study, Anneli
Aejmelaeus goes some way to doing all of this. I quote her conclusions, with which
I concur for the most part.

[T]he frequent use of ἐλπίζω in the Greek Psalter is in several respects unique
among the books of the Septuagint. The verb was clearly a conscious choice by the
translator—this is implied by its frequency—and it was used in a very pregnant
sense. . . . In the Psalms it is obvious that ἐλπίζω is used with positive connota-
tions: “to expect good things from,” “to put one’s hope in.”38

This is to say that in the Greek Psalter ἐλπίζω carries the sense located in domain
25 of Nida and Louw’s taxonomy, attitudes and emotions, subdomain D, hope, look
forward to. As such it is a close neighbor of the verb ἀπεκδέχομαι, in the sense to
await eagerly or expectantly for some future event. By downplaying the prospective
and affective senses of ἐλπίζω, Muraoka and Zimmerli do violence to the plain
meaning of the Greek Psalter. For the translator clearly favored the word as a match
for a number of Hebrew verbs that characterize the relationship of the psalmist to
YHWH. By glossing it as “trust” we obscure this fact and read the Greek as if it
were Hebrew. In so doing, we lose sight of a truly remarkable development—not
a change in semantics but a change in piety: the advent of a distinctive religious
language.39 This is an idea that I would like to develop further.

38. Anneli Aejmelaeus, “Faith, Hope, and Interpretation: A Lexical and Syntactical Study
of the Semantic Field of Hope in the Greek Psalter,” in Studies in the Hebrew Bible, Qumran, and
the Septugint: Presented to Eugene Ulrich (ed. P. W. Flint, E. Tov, and J. C. Vanderkam; VTSup 101;
Leiden: Brill, 2006), 360–76, here 370.
39. Similar conclusions are drawn by Anneli Aejmelaeus, “Levels of Interpretation: Trac-
ing the Trail of the Septuagint Translators,” in On the Trail of the Septuagint Translators: Collected
Essays (rev. and exp. ed.; CBET 50; Leuven: Peeters, 2007), 295–312, here 303.

HiebertA.indd 50 8/30/2010 7:52:14 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 51

5. A Rhetoric of Hope
What is the significance of the translator’s use of ἐλπίζω? I have stressed that we
cannot read too much into any given context in which the word is used since it is a
default match for three Hebrew verbs. But as Aejmelaeus points out, these matches
are themselves significant. The translator has consistently used ἐλπίζω where he
might well have used other Greek verbs. Furthermore he has favored the noun
ἐλπίς, “hope,” in a similar manner.
This preference for ἐλπίζω and ἐλπίς is of decidedly thematic import, yet
we must stop short of speaking of a theme of hope in the Greek Psalter. It is not a
compositional text. Again we come back to its textual linguistic makeup. Given the
translator’s reliance on formal equivalency, he was not in a position to introduce
and develop themes.
That said, there is no denying that deliberate choices on the translator’s part
give rise to a Greek liturgical language in which the dominant note is one of expec-
tation. I think, then, that we might without contradiction talk about a rhetoric of
hope. By rhetoric I simply mean language calculated to have a persuasive effect.
The verbal makeup of the Greek Psalter, I would suggest, is such as to make a
certain impact on the piety of the implied reader, namely, to focus it on a desired
future.40
A signal feature of this rhetoric is the frequent occurrence of ἐλπίζω in the
imperative mood, construed with ἐπί and either κύριος or θεός, as in “Hope in
God!” (41:6, MT 42:6) or “Let Israel hope in the Lord!” (130:3, MT 131:3). To bor-
row a happy phrase from Walther Zimmerli, the implied reader of these verses,
the one who turns to the Lord, is cast as a “creature of expectation.”41 We cannot
say much more. Dealing as we are with a formal translation, one that leans heav-
ily on the form of the Hebrew source, we simply do not know how the translator
understood the object of Israel’s hope, the desired future. As I have stressed, there
was no opportunity for him to develop it as a theme. The Greek Psalter is jealous
of its secrets, and withholds any further insight into the matter. Such is the nature
of interlinear translation.
We cannot fathom the mind of the translator, but having identified a rhetoric
of hope in the text, we might go on to describe its larger literary context, and per-
haps even hazard a few guesses as to the sociocultural milieu in which it arose. The
goal here would be to provide a thick description of the rhetoric, that is, a descrip-

40. It is important to appreciate that the Psalter’s rhetoric of hope stands in stark contrast
to typical treatments of the theme in Hellenistic literature. See Despland, Education of Desire,
281. The tendency of Greek authors is to counsel against hope, which, bound up as it is with
desire, is delusional, and a temptation to be resisted. Hence the language of the Psalter is decid-
edly marked in this respect.
41. Zimmerli, Man and His Hope, 30. Note, however, that Zimmerli is speaking in refer-
ence to the Hebrew text.

HiebertA.indd 51 8/30/2010 7:52:14 PM

 52 “Translation Is Required”

tion of the cultural context in which it was meaningful to the implied reader of the
text.42 The semantic horizons of such an inquiry would ideally remain focused on
textual relations rather than reconstructed histories.
Of course, the Psalter’s rhetoric of hope is not merely of literary significance;
it is arguably of theological import as well. The question thus arises as to where
this aspect of the text might fit in a thick description. In the next section I shall
consider this problem. Having begun with Barr’s early work on semantics, it is fit-
ting to close with a brief glance at biblical theology, a topic with which he wrestled
in his later years.

6. Toward a Theology of Hope

There have in recent times been many calls for a theology of the Septuagint, but
we must proceed with caution. Biblical theology, James Barr warned us, is itself
an inherently problematic undertaking. If we define theology as a reflective activ-
ity in which the content of religious expressions is abstracted, reformulated, and
subjected to discussion, then as Barr observed, much of the Bible does not have
this character: “Religious opinions, expressions and aspirations, however strongly
expressed, are not thereby theology.”43 So, without denying that they imply theol-
ogy, the question must be raised whether, properly speaking, biblical texts contain
theology.
Yet Barr himself readily admits that a movement can be traced whereby the
biblical religion reformulates itself as something closer to theology. One major
catalyst, he suggests, was life in the Hellenistic world. He agrees with Gerhard
Ebeling’s proposal that, above all, it was contact with Greek thought and language
that brought biblical religion to the threshold of “theology.”44 On this view, apoc-
ryphal books such as the Wisdom of Solomon contain theology. What about the
Greek Psalter?
It is indeed tempting to think that the process of translation from Hebrew
into Greek involved the translator in theological inquiry, which, in turn, is itself
reflected in the translation. And with books such as Proverbs or Job this may
in fact be the case, for in both instances the translators reformulate the Hebrew
source. The problem in talking about a theology of the Greek Psalter is the model
of translation underlying it. Here, as I have stressed, the translator has adopted a
method of formal or isomorphic translation; he is very tolerant of interference

42. The term “thick description” originates in the work of the Oxford philosopher Gilbert
Ryle. Its use in cultural anthropology stems from Clifford Geertz, “Thick Description: Toward
an Interpretive Theory of Culture,” in The Interpretation of Cultures: Selected Essays (1973; repr.,
New York: Basic Books, 2003), 3–30.
43. Barr, Concept of Biblical Theology, 249.
44. Ibid.

HiebertA.indd 52 8/30/2010 7:52:14 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 53

from the source.45 The result is what we might call fractured discourse: it often
lacks both coherence and cohesion. The translator frequently declines to provide
us with a well-formed text—again and again he falls back on the structure of the
Hebrew parent. So it is very difficult to identify theological propositions that may
credibly be attributed to him.
But what if we construe the task of biblical theology as primarily exegeti-
cal? In other words, can we engage in a theological exegesis of the Greek Psalter?
This, as Barr suggests, would involve relating specific biblical texts to theological
themes.46 We need not thereby impute such themes to the author, but in identify-
ing them we would hope to gain some insight into the theological meaning of the
text for its implied reader.
A potentially fruitful theological background against which to read the Greek
Psalter’s rhetoric of hope, I would suggest, is the Torah piety that had entered Jew-
ish wisdom literature by the early Hellenistic period. James Luther Mays describes
it as follows:

The two primary problems with which it lived were wickedness in self and society
and the arrogance and power of the nations. The questions with which it wrestled
were the incongruity of conduct and experience and the hiddenness of the pur-
pose of God in history. Its way was faithfulness through study and obedience and
hope through prayer and waiting.47

The origins of Torah piety are not clear, though it has been suggested that it
emerged under Hellenistic influence, reflecting the Socratic view that the soul is
ennobled and made virtuous through the pursuit of knowledge.48 Stephen Geller
notes that such piety is in one sense profoundly inward-looking: subjective states
such as waiting and expectation are dominant themes.49 While there is no single
Hebrew word with the same semantic range as ἐλπίζω, we find that a variety
of linguistic resources are mustered to express a personal piety characterized by
hope. This is particularly true in Hebrew Ps 119,50 one of the most fully developed

45. Pietersma, “Psalms: To the Reader,” 542–44.

46. Barr, Concept of Biblical Theology, 251.
47. James Luther Mays, The Lord Reigns: A Theological Handbook to the Psalms (Louisville:
Westminster John Knox, 1994), 134.
48. Shaye J. D. Cohen, From the Maccabees to the Mishnah (Louisville: Westminster John
Knox, 1987), 72.
49. Stephen Geller, “Wisdom, Nature and Piety in Some Biblical Psalms,” in Riches Hidden
in Secret Places: Ancient Near Eastern Studies in Memory of Thorkild Jacobsen (ed. I. Tzvi Abusch;
Winona Lake, Ind.: Eisenbrauns, 2002), 116.
50. Leslie C. Allen (Psalms 101–50 [WBC 21; Nashville: Thomas Nelson, 2002], 183)
argues that, given the presence of Aramaisms and terms characteristic of late or postbiblical
Hebrew, a postexilic date for Ps 119 is inescapable.

HiebertA.indd 53 8/30/2010 7:52:14 PM

 54 “Translation Is Required”

expressions of Torah piety.51 As Walther Zimmerli has shown, it is dominated by

the theme of hope.52
Meditation on God’s Law, Zimmerli points out,53 constitutes the Hebrew
psalmist as a “creature of expectation.” He bids God, “Remember your word to
your servant, in which you have made me hope (yntlxy)” (v. 49). The divine word
that speaks to him and guides him also opens him to the future; it awakens his
expectations and promises help.54 “My soul languishes for your salvation,” he
declares, “I hope (ytlxy) in your word” (v. 81). This hope is not merely private; it
is a source of inspiration for others. “Those who fear you shall see me and rejoice,
because I have hoped (ytlxy) in your word” (v. 74).
What is particularly interesting about Ps 119 is that it constructs what we
might call a soteriological context for Torah piety. The ideal situation proposed at
the beginning of the Psalm, that prosperity is guaranteed to the righteous, is chal-
lenged throughout.55 But the psalmist can persevere in his adherence to Torah,
even in the face of personal failure and persecution, because God’s word nourishes
a hope, that is, the promise of salvation. Whether this soteriological context has an
eschatological horizon, as Mays would argue,56 is unclear. It is certainly a hope for
some kind of renewal of life.
I would suggest that the recourse of Jewish wisdom to hope as the characteris-
tic attitude of the Torah-centered sage is reflected in the verbal texture of the Greek
Psalter.57 In a third-century Hellenistic context, it is altogether possible that this

51. Hans-Joachim Kraus (Psalms 60–150: A Commentary [Minneapolis: Fortress, 1993],

414) describes Ps 119 as “a collection of statements of the individual Torah piety of postexilic
times that originated from elements of the study of Scripture, of Deuteronomic theology, of cultic
Torah instruction of an individual and of the stimulus of wisdom teaching.” Walter Bruegge-
mann (Reverberations of Faith: A Theological Handbook of Old Testament Themes [Louisville:
Westminster John Knox, 2002], 219) characterizes the psalm as “a summary of Torah piety.”
52. Zimmerli, Man and His Hope, 29–31.
53. Ibid., 30.
54. Ibid.
55. Dianne Bergant, Israel’s Wisdom Literature: A Liberation-Critical Reading (Liberation-
Critical Reading of the Old Testament; Minneapolis: Fortress, 1997), 63.
56. Mays, Lord Reigns, 134. Cf. Jerome F. D. Creach, The Destiny of the Righteous in the
Psalms (St. Louis: Chalice, 2008), 140: “Together Psalms 118 and 119 put torah piety in an escha-
tological context. Hope for the coming kingdom, for God’s reign, is a hope for realization of the
truth that is proclaimed in the torah psalms.”
57. Jerome F. D. Creach (Yahweh as Refuge and the Editing of the Hebrew Psalter [JSOTSup
217; Sheffield: Sheffield Academic Press, 1996], 33) likewise finds it interesting that, while the
Greek translator evidently knew the difference between x+b and hsx, he nevertheless rendered
both by ἐλπίζω when they refer to reliance on YHWH. Creach goes on to suggest that the trans-
lator may have understood these terms in reference to a “refuge piety,” evident in late Hebrew
psalmody (and the focus of Creach’s study), which viewed personal reliance on YHWH as the
supreme virtue. Since, according to Creach, “the most concrete way of expressing such reliance
was in the study of Torah” (48), it is clear that what he is describing is a special case of Torah

HiebertA.indd 54 8/30/2010 7:52:14 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 55

hope for renewal was understood eschatologically, perhaps in terms of a resurrec-

tion of the dead, as Joachim Schaper and others have suggested, though one can-
not say for sure.58 It is true that members of the Orphic movement had long distin-
guished themselves from other Greeks in claiming to have a hope.59 For Orphism,
the gods are good, and pious men and women may hope to fare well in the afterlife,
even though they now suffer on account of their virtue.60 Plato had reconstrued
this hope as the reasonable expectation of the virtuous and, in particular, of those
who embraced the philosophical life. In his portrait of Socrates, the philosopher
expresses the self-assurance of the man who is of good hope;61 that Socrates has a
hope in the afterlife is stressed.62 This form of hope is encountered in 2 Maccabees,
where it is refigured as the hope of resurrection.63 Just such a hope may well be the
key to understanding the rhetoric of the Greek psalmist. But, as I have said, the
eschatology of the Greek text eludes us; the semantics of the translation are such
that we may never know.

piety. This being so, there is a pleasing congruence between his reading of the Greek Psalter and
my own.
58. Joachim Schaper (Eschatology in the Greek Psalter [WUNT 2/76; Tübingen: Mohr Sie-
beck, 1995], 50) discovers in Ps 15(16):9–10 “one of the first, if not the first” instance of the prom-
ise of “personal, physical resurrection.” In support of this claim, Schaper appeals to the rendering
of the prepositional phrase x+bl (“in security”) by ἐπ᾽ ἐλπίδι (“upon hope”), and tx#$ (“pit”)
by διαφθοράν (“destruction or physical corruption”). Yet the Greek expressions do not strike
one as sufficient in themselves to carry the interpretative weight Schaper wishes to place on them.
See the discussion in Cameron Boyd-Taylor, “In a Mirror, Dimly: Reading the Septuagint as a
Document of Its Times,” in Septuagint Research: Issues and Challenges in the Study of the Greek
Jewish Scriptures (ed. Wolfgang Kraus and R. Glenn Wooden; SBLSCS 53; Atlanta: Society of
Biblical Literature, 2006), 15–31.
59. Despland, Education of Desire, 132.
60. Ibid., 132–33.
61. Ibid., 132.
62. E.g., Phaed. 63b4–c7. “I not only do not grieve, but I have great hopes [εὔελπις] that [AQ: Since this
there is something in store for the dead, and, as has been said of old, something better for the isn’t LCL, please
good than for the wicked” (63c; Benjamin Jowett). For a stimulating philosophical discussion of cite in full.]
Socrates’ hope for an afterlife as expressed in the Phaedo (a hope shared by his companions in the
dialogue), see Peter J. Ahrensdorf, The Death of Socrates and the Life of Philosophy: An Interpreta-
tion of Plato’s Phaedo (Albany: State University of New York Press Press, 1995), 35–37.
63. See 2 Macc 7:14 where reference is made to τὰς ὑπὸ τοῦ θεοῦ προσδοκᾶν ἐλπίδας
πάλιν ἀναστήσεσθαι ὑπ’ αὐτοῦ. In 2 Macc 7:20 the phrase διὰ τὰς ἐπὶ κύριον ἐλπίδας
implies the expectation of resurrection. Jonathan A. Goldstein (II Maccabees: A New Translation
with Introduction and Commentary [AB 41A; Garden City, N.Y.: Doubleday, 1983], 285) observes
that no educated Greek reading the characterization of Eleazar at 6:18–31 could miss his resem-
blance to Socrates: both hold that it is better to go to the underworld maintaining obedience
to the laws (6:23; Crit. 54b–d); both hold that though one may escape human punishment, one
cannot escape divine punishment (6:26; Apol. 39a-b); both trust in supernatural judges (6:26, 30;
Apol. 41a).

HiebertA.indd 55 8/30/2010 7:52:15 PM

 56 “Translation Is Required”

Nevertheless, through his use of ἐλπίζω and ἐλπίς, the translator of Psalms
has undoubtedly made hope in a God who saves the leading motif of the Greek
Psalter. This fact has interesting implications for theological anthropology. The
dominant note struck in the Hebrew Psalter, observes Brevard Childs,64 is that of
human frailty and vulnerability. However, in spite of this emphasis, an important
dimension of being human for the psalmist is our capacity to hope. As we have
seen, this dimension of the text is foregrounded in the Greek version. This may
reflect in part the Hellenistic assumption that humanity is characterized preemi-
nently by its reliance on hope. As mortals we are denied certainty regarding the
future—we can only hope.
What distinguishes the picture of our common condition found in the Greek
Psalter, if I may put the matter theologically, is its conception of hope as the yearn-
ing for a life renewed by God. I use the word “conception” advisedly. We are no
longer talking about the lexical meaning of ἐλπίζω or ἐλπίς, but the meaning of
hope as such. Yet a concept may subtly alter the semantics of the words used to
express it, especially in terms of their supplementary features, that is, those that are
relevant only in certain contexts or that play a primarily connotative or associative
role. That this was the case for ἐλπίζω and ἐλπίς is, I would suggest, not unlikely
in Philo, and almost certain in the Apostle Paul.
In Quod deterius potiori insidiari solet (That the Worse Attacks the Better),65
Philo takes up a distinction made in Stoic anthropology between the πάθη, or
“passions,” which dog the wretched man, and the corresponding “states of bless-
edness,” the εὐπάθεια, enjoyed by the one who seeks after virtue. Yet he departs
from Stoic theory in including ἐλπίς as one of the εὐπάθεια. The expectation of
achieving virtue produces hope, writes Philo, “that food of souls which makes us
cast away hesitation and attempt with hearty alacrity all noble deeds.” Later in the
same treatise Philo characterizes the human being as a soul so constituted as “to
hope on the God that really is [τὸν ὄντως ὄντα θεὸν ἐλπιζούσης]” (139). I do
not want to attribute Philo’s anthropology to the Greek Psalter, but clearly both
place a theological value on our capacity to hope.
I close with Paul and, in so doing, end where I began, for in one respect Spicq
was right. Paul was indeed one of the principal heirs of the language, the lexicon,
and the faith expressed by the Greek Psalter.66 Spicq’s mistake was to infer that this
involved a fundamental change in the Greek lexicon, whereby Greek words had
taken on Hebrew meanings. New Testament ἐλπίς, Spicq writes, “is sure and cer-
tain by virtue of its semantic origin in the LXX, where it means essentially having
confidence, being assured.”67 On the contrary: what Paul inherited was a Greek

64. Brevard S. Childs, Biblical Theology of the Old and New Testaments: Theological Reflec-
tion on the Christian Bible (Minneapolis: Fortress, 1993), 573.
65. Philo, Det. 120 (Colson and Whitaker, LCL).
66. Spicq, Theological Lexicon, 1:486.
67. Ibid., 490.

HiebertA.indd 56 8/30/2010 7:52:15 PM

 Boyd-Taylor: Semantics of Biblical Language Redux 57

meaning put to the service of a distinctive theologoumenon, hope in the God who
saves. The prospective and affective elements of ἐλπίς/ἐλπίζω remain. To read
Paul otherwise, I would suggest, is to lose sight of the “tension and emotion” in
his thought.68 As Herman Ridderbos writes, Pauline hope directs itself toward the
invisible things of the future that are eternal and derives its strength from them.69
To quote the apostle: “For in hope we were saved. Now hope that is seen is not
hope. For who hopes for what is seen? But if we hope for what we do not see, we
wait for it with patience” (Rom 8:24–25).

68. See Herman Ridderbos, Paul: An Outline of His Theology (Grand Rapids: Eerdmans,
1975), 249. For Paul, hope is indissolubly bound up with faith: “[O]n the other hand, they both
represent no less the provisional revelation of the new life and of the new man, and it is striking
how much the apostle alternatively places the emphasis on the one and then the other. In this way
there enters into the concept of faith the strong tension and emotion that are so characteristic of
the manner in which Paul, often in the most personal passages in his epistles, gives expression to
his own experience of faith.”
69. Ibid., 248.

HiebertA.indd 57 8/30/2010 7:52:15 PM