Lexical Semantics and Lexicography

Part 2 of the series Observations from a Linguistic Spectator: An Annual Report.

For part 1, see here.

A Semiotic Framework

I’m going to cheat right away in this first post in this series that discusses an actual linguistic subject – quite a bit of what I discuss here isn’t that new to me. So it’s not really part of the honest annual report of new insights that I promised. Still, I feel like I have to include these considerations because in my experience this is perhaps the single issue where publications in biblical studies fail most often.

I say this as someone who has done quite a bit of research on the meaning of some Greek words – even a whole monograph on θριαβεύειν – but who is at the same time very aware of the fact that he’s working within a quite simplistic framework. (If I’ve made a somehow innovative contribution it’s on the question of how we are supposed to choose between several available meanings in a dictionary in concrete occurrences of a word – a task where biblical scholars often make mistakes in my opinion.)

In fact, most basically, I’ve just tried in my own work simply to explicate the semiotic triangle that probably most of you know:

The original triangle from Odgen and Richards.

The symbol, the Greek word in question, stands for a concrete thing in the real world. But it does so only in a mediated way, namely by symbolizing/expressing a concept, a mental category, which is in turn represented by the symbol. It’s the concept that refers to the actual referent, which in turn is reflected in the concept.

This is certainly not advanced semiotics. And still, it’s astonishing how often studies on the NT fail to do justice to this simple scheme. And let’s be totally clear: A mistake with respect to differentiating properly between these levels is a fundamental category mistake that will render the whole work completely wrong at worst and almost not usable (because in need of constant reinterpretation) at best. Regardless of how keen you are on incorporating linguistic tools in your research, I hope we can agree that this is not some unnecessary abstract hair-splitting but something that is of the uttermost importance for sound exegetical work.

Explicating Barr’s Criticism

Many of James Barr’s criticisms in The Semntics of Biblical Language – arguably the most important book on the subject in the last century – can be reformulated in these terms:

  • Perhaps his most important point (though this is seldomly reflected in works that paraphrase him) was his rebuttal of the Sapir-Whorf hypothesis that had been common among biblical scholars, who claimed that the languages Hebrew and Greek were determinative for different world views. Barr argued forcefully not only that every language is capable of expressing the same thoughts but even that concepts like ‘truth’ were universal, regardless of how that was reflected on the level of the language.
  • He accused Kittel’s T(h)WNT/TDNT of constantly confusing the levels of words and concepts. While the editors claimed to produce a “Wörterbuch,” they were actually discussing concepts – which were far more complex than the lexical meaning of the words in question. He spent a lot of energy showing how this confusion was at least facilitated by an inconsequent usage of the word “Begriff” – ranging from everything between it being an equivalent to “word” on the one hand and to “concept” on the other hand (or both at the same time at some places).
  • Barr showed that often the authors succeeded in sketching these far-reaching concepts by actually discussing referents – and confusing these with words. So, for example, the way the love of a specific entity manifested – e.g. ἡ ἀγάπη τοῦ θεοῦ – was read into every occurrence of the word ἀγάπη (the well-known “illegitimate totality transfer”), even though many of these features had nothing to do with the lexical meaning. In a further step, this enriched account of the “meaning” of ἀγάπη was then presented as an exhaustive account of the concept of ‘love’ (see #2).

Continuing Category Mistakes and Problematic Nomenclature

At least in English-speaking scholarship these criticisms are well-known, at least in the form of the summaries that appear in Carson’s Exegetical Fallacies and Silva’s Biblical Words and Their MeaningIn German publications, however, there still is at times an astonishing ignorance when it comes to these points. Some scholars still talk about words as if they were concepts and about concepts as if they were words.

Take, for example, the impressive volume Glaube: Das Verständnis des Glaubens im frühen Christentum und in seiner jüdischen und hellenistisch-römischen UmweltIt contains a wide array of well-researched pieces and is very valuable in bringing together experts from different fields and an incredible number of references to primary sources (I know because I prepared the index). Judging from the title and the subtitle, it’s a conceptual study, i.e. just what Barr had encouraged – especially since it seems to be marked specifically as such, i.e. insofar as apparently no Greek word is needed to describe the task on the cover. That being said, some (!) of the essays seem to explore rather “das Verständnis von πίστις,“ i.e. the question of which meaning can be ascertained for a specific lexeme in a given corpus. Of course there is nothing wrong about that question in itself. And to be sure, any exploration of the concept of ‘faith’ will take into account the noun πίστις – or rather, the passages that contain it. But while this is a necessary condition for a faithful account, it certainly isn’t a sufficient one. As Barr emphasised over and over again: the meaning of whole paragraphs (I would say: texts) is relevant for such an endeavour and they might not contain a single one of these words. I want to reiterate that just a few of the authors are guilty of apparently not being aware of this. We should also note that the editors Jörg Frey and Benjamin Schliesser both explicitly quote James Barr (on illegitimate totality transfer and the distinction between a Hebrew and a Greek mindset). At the same time but Barr’s urgent plea to end the confusing usage of the word “Begriff” is certainly not implemented enthusiastically by many of the authors, with the word family appearing over 600 times… And you still read about “Begriff” in the sense of ‘concept’ along with talk like this (to adduce just one entirely random example):

“Die hebräische Sprache kennt eine Reihe von Begriffen für das Vertrauen des Menschen auf Gott als den Schöpfer, Erhalter und Retter in der Not… ”

Apparently, here the author suddenly switches to the meaning ‘word’ for “Begriff.” To be fair, some authors in this volume (for example Stefan Krauter) seem to use this word exclusively with this meaning. That’s certainly less confusing. Still, we should note that the Wörterbuch der deutschen Gegenwartssprache lists only meanings that are clearly located on the conceptual level. That was 1967. Admittedly, the Duden now also lists the meanings ‘Ausdruck, Wort.’ Bute note that it’s still marked specifically as being colloquial!

But beyond that, I am not even sure whether this is a case of real polysemy or whether a single understanding of the word is presupposed that is so undifferentiated that it does not even make a distinction between a dyadic semiotic demarcation between sign and content. For in this specific case the author immediately goes on to say:

“Davon noch einmal zu unterscheiden ist der spezifische Gottesglaube, der im Sinne eines grundsätzlichen Trauens bzw. Für-Wahr-Haltens auf Gott als Gegenstand gerichtet ist und in der Forschung mit der Wurzel אמן hif. in Verbindung gebracht wird. Die Bedeutung dieser Wurzel für das Glaubensverständnis wird auch daran ersichtlich, dass sie über das griechische Verb πιστεύω in der Septuaginta auf den neutestamentlichen Glaubensbegriff fuhrt.”

While the formulation that a specific lexeme is “connected” with a certain conception of ‘faith’ might perhaps be interpreted as reporting an opinion that is not shared, I don’t know what to do with the move from the certainly conceptual “Glaubensverstänis” in the OT to the (I guess equally conceptual?) “Glaubensbegriff” in the NT, as it is mediated through nothing but a specific translation equivalent in the LXX. That’s not only inconsequent or confusing usage of terminology, it’s presupposing a concept of the word “Begriff” that is not tenable against the background of the semiotic framework sketched above. It’s exactly the kind of thing that would make Barr turn in his grave. And it’s no exception. At many places, you’ll find “Begriffsgeschichte” in the sense of a conceptual history directly and without any explanation connected with an overview over diachronic lexical developments. I was especially surprised to at times notice a direct inference from lexical realization to the shape of concepts even if the author apparently made a distinction between the levels of words and concepts:

“Allein schon die Zahl weiterer, dem Glaubensverständnis bei Philo im weiteren Sinn zuzuordnenden – von Paulus aber nicht, nicht in dieser Weise oder selten verwendeten – Begriffe und Syntagmen deutet an, dass wir es bei Philos Glaubensverständnis nicht nur mit einem komplexen und auf vielerlei Weise ausgedruckten Sachverhalt zu tun haben, sondern auch im Ganzen mit einem Paulus nicht einfach vergleichbaren theologischen Konzept.”

In my mind these cases constitute a serious failure to take into account Barr’s criticism. At least at times it renders whole contributions totally questionable, because as a reader you are not able to translate the claims into a linguistically sounder framework. If you think that assessment is too harsh, I encourage you to try translating such a piece into English yourself. Translators of works in biblical studies will tell you, that “Begriff” is the single most annoying term they encounter. (In fact, I just checked and Wayne Coppins calls it indeed his (probably) “least favorite German word.”) Just look at what happened to the German Theologische Begriffslexikon zum Neuen TestamentIt was mistaken even by the editors of the English translation to be a concept-lexicon and has been in use as such for decades, as a kind of implementation of Barr’s suggestion, while it is demonstrably none of that! (If that sounds just too unbelievable to you, check out my detailed analysis here.)

Another problematic tendency in German-speaking scholarship is to still use specific Greek terms when one is referring to concepts much larger than the semantic content of that single lexeme. To give one example, Manuel Nägele is currently doing a doctoral thesis with Jörg Frey with the title “Paulus und der νοῦς: Eine Untersuchung zur paulinischen Anthropologie auf dem Hintergrund hellenistischer und hellenistisch-jüdischer Konzeptionen.“ The subtitle is pretty clear when it comes to what the researcher wants to do: he’s interested in “anthropology” and different ancient “conceptions.” (Cf. also the conference associated with the project with the title “Der νοῦς bei Paulus im Horizont griechischer und hellenistischjüdischer Anthropologie.”) However, when I first heard a presentation of this project, I had the following criticism (note that it might no longer apply, since this was a presentation at a very initial stage – and in fact I think it’s great if such a setting of the course can be corrected early on): Why is there mention of νοῦς in the title? It’s a noun that occurs over 65.000 times in the TLG corpus but just 21 times in the Pauline letters. Thus, it can hardly be the goal to assess the meaning of the lexeme in a lexicographical way on the basis of this small corpus. Assuming that it’s a polysemous word, one could of course try to find out, which of the meanings are adopted at the 21 places. That’s perhaps a a good research topic in itself, but it contributes almost nothing to what subtitle implies. So, one wonders, perhaps the Greek word νοῦς is simply used instead of the German word “Verstand”? It seems so. The author is interested in how Paul’s conception of the mind differs from other thinkers of his time. An interesting question indeed. But why then not just say so? One could add a clarifying “Konzept des Verstandes” – even though I think it’s not necessary because it’s clear that Paul didn’t write in German. In any case that would be better than muddling the water by using a Greek expression. I assume the word made its way into the title, because the author selects those passages for his analysis where the word occurs. That is of course totally fine as a way of making the first step towards a full conceptual study feasible. But note that the author is then analysing what the passages that contain νοῦς say about the concept of the mind – neither what they contribute to Paul’s understanding of the word νοῦς nor what the word νοῦς contributes to the larger concept. In Barr’s critique, these are totally different things and they need to be kept separate. To be sure, such a problem in the label doesn’t have to have detrimental effects on the project – but at the same time we have to maintain, that of course it’s only a full conceptual study if it does not focus exclusively on the passages that contain νοῦς, even less so of course on what the lexeme itself “means” in these passages.

There is one tendency in NT publications that I find even more problematic – the transcription of Greek words in project titles, such as “Pneuma/Pistis bei Paulus.” Every time I read something like that, I am at a loss. Sometimes the cover text helps in explaining the general goal of the study, sometimes even reading the whole book doesn’t help, because the title is symptomatic of a lack of necessary differentiation. In my opinion, the use of Greek words in order to refer to a concept is strange enough (if you aren’t writing in Greek) and requires efforts of interpretation on behalf of the reader. But this is even worse: Perhaps the author wants to indicate that the concept diverges so much from anything our world view contains that it’s necessary to find a label that symbolizes this remoteness? Or is it just the case that the publisher’s font doesn’t allow for Greek characters on the book cover? Not too seldom, you then even encounter “Begriff” in the subtitle… Eckstein even manages to put both in the main title: “Der Begriff Syneidesis bei Paulus.” And just when you begin to think that perhaps you are holding a lexical-semantic work in your hands, the subtitle sends you on a longer journey of figuring out what the book is about: “Eine neutestamentlich-exegetische Untersuchung zum ‘Gewissensbegriff.’”

The Prospects and Limits of a Sound Theoretical and Methodological Framework

Keeping distinct referents, concepts, and words and having this reflected in the terminology of one’s writing not only makes sure that other scholars are able to interact with the work constructively, it also helps to stay on top of things during the research itself. For example, I’m currently looking into the semantics of μονή, given that N. T. Wright and Markus Bockmuehl have clashed over the question of whether it implies an only temporary resting place in John 14. I’m looking into different things here: I want to know a lot about ancient institutions of temporary and permanent dwelling. That’s the kind of information an encyclopaedia like the Pauly offers. Such entries are relevant for but will be more exhaustive (or at least have different emphases) than the concepts that constitute lexical meanings. And of course I’m also interested in how these concepts are expressed lexically. If I didn’t differentiate sufficiently between the different semiotic categories, I might easily fall prey to mistaking my description of an ancient inn as the lexical meaning of μονή (or κατάλυμα for example).

So keeping apart word, concept, and referent – and having this reflected in a consistent terminology – really goes far in making the actual research output linguistically sound and accessible to others in the field. It is not, to be sure, a guarantee that the results will be correct. For example, Breytenbach in an article argued that transitive θριαμβεύειν only implies the celebration of a triumphal procession over the person referred to in the direct object. The presence of that person in the procession – as a captive who is represented to the crowd – is said to be “a matter of reference, and reference, as we all know, has to do with language use and should not be confused with lexical sense” (p. 262). The problem is that the “language use” demonstrates that authors regularly assumed the verb to express the causation of movement of another entity in the triumphal procession by the triumphator (see more here). Still, the fact that Breytenbach makes his claim in a way that keeps basic distinctions clear allows for his study to be reproduced and his results to be tested without first having to translate his whole analysis in a sound framework in order to decide which parts are actually relevant. And this is far from being a matter of course! Most other proposals for the lexical meaning of θριαμβεύειν are actually discussing concrete ancient rituals and other realities of life without even touching on the question of whether the verb is ever used as a “symbol” for these referents, let alone whether it is also a plausible lexical choice.

Problems with Barr’s Concept of Concept

So far, I’ve mostly repeated what I’ve also already written in some form in other places and what I know is frustrating to many colleagues, some of whom know much more than I do about lexical semantics and lexicography. In fact, if you have read Barr’s book from 1961 (or the German translation from 1965) almost nothing will probably sound new to you. But I hope that my illustrations have at least proven the point that these distinctions are still not a matter of course in biblical scholarship and that it is thus appropriate to repeat them from time to time.

Now I’ll turn to something that was indeed new to me and that I only learnt recently. Or at least it was only recently that I realized the full implication of this insight. As I will explain in what follows, there are good reasons to believe that Barr was mistaken in quite a few of his basic assumptions – and that complicates some of the matters that we’ve just discussed immensely!

There’s one tension in Barr’s work that I had noticed even when reading it for the first time. On the one hand, he insists that, regardless of lexical options in a language, in the mind of its user concepts such as ‘truth’ will be the same as in other speech communities. On the other hand, he’s actually supportive of the implicit goal of the TDNT of producing a “concept history” – he just doesn’t want it to come together through the misuse of word studies. He even writes his own “Begriffsgeschichte” of the word “Begriff,” demonstrating that not everybody shares his understanding of the term, given that the meaning ‘concept’ presupposes that it’s even possible to make such a distinction. To me, it has always been difficult to see how the two positions relate to each other. Clearly, if you think that it’s worthwhile to analyse whole passages to see what they have to contribute to a specific concept and if you do so with texts of a specific corpus such as the New Testament or early Christian literature, you presuppose that there is variation on the conceptual level. Only because of this variation does it make sense to incorporate different voices from the same tradition (unless you use them only to supplement gaps in other accounts) and specifically to be interested in the specific shape of concepts in a particular tradition in contrast to other traditions. Barr does not do a good job in my opinion in explaining what he regards to be the constant core of a given concept and how and to what degree variations of it can exist. In my opinion, John Barclay’s work on different “perfections” of the concept of ‘grace’ is by contrast a step in the right direction.

But there are other problems in Barr’s work that go beyond the lack of explication. They also, however, have to do with Barr’s problematic view of what concepts are. I first realized that something was wrong with Barr’s account, when I read more cognitive linguistic literature. The way we conceptualize things has a lot do with how we experience them. By expressing these concepts in language, our experience becomes manifest on a lexical and grammatical level. For example, prepositions can be understood to a large degree with reference to our embodied experience that we have in physical space. Prepositions are not determined by our experience, but they are also not arbitrary. Now, in some sense this supports what Barr had written about the universal nature of concepts: we all share the experience of the same physical laws and we will thus have similar concepts of ‘above’ and ‘below’ and will have developed strategies of expressing these relations in our languages. On the other hand, experience varies not only individually to some degree but also collectively between times and cultures. If experience is so important on the conceptual level and if language is (amongst others) a tool of expressing these thoughts, language will have different shapes due to different conceptions and be ultimately influenced by experience. Note that this does not vindicate the Sapir-Whorf hypothesis. But it shows that there is an organic relationship between the three corners of the semiotic triangle. In other words, the conceptual level does not float around in lofty heights above the level of expressions.

The Structuralist Heritage of Barr’s Critique

To put it differently, words “have a nontrivial relationship with the external world and human knowledge.” That’s how Mike Aubrey (follow his blog here) puts it in his piece “Linguistic Issues in Biblical Greek” in Linguistics and Biblical Exegesis (p. 182; I would strongly encourage you to read at least pp. 173-183 on semantics and lexicography; he had posted a portion of it on his blog). This misconception of Barr has to do with his structuralist presuppositions. In structuralism, the meaning of words is determined by their relationship to other words – everything gains relevance from its place within a network of signs of the same code. Only after having read Aubrey’s piece did I realize how much I had been influenced myself by the structuralism of Louw and Nida and how much I had supposed that definitions are meanings.

Now, don’t get me wrong: I still believe definitions are crucially important. As John Lee has shown in his History of New Testament Lexicographythe use of glosses instead of more elaborate definitions (and their uncritical handing-down through generations of lexicons) has had a horrible impact on how the meaning of Greek words in the NT is presented to exegetes. Of course, it’s often unproblematic to work with glosses if the Greek and German/English/French/etc. words express concepts that overlap significantly. Almost always, however, their boundaries don’t fall together and if you use a gloss for a Greek word you basically assign the boundaries of the gloss in the target language to that word – causing incredible confusion at many places.

That being said, the insights of cognitive linguistics also show that the meaning of words is much more than just what we manage to capture in definitions. This also means that componential analyses are only part of the story. I myself had quite heavily relied on this framework, which assumes that you can determine the meaning of a word by comparing it to how it relates to other words in the same semantic domain with reference to a set of features. I used this illustration from Cotterell and Turner in Paul’s Triumph (p. 14):

 Seating for one person, vs. more than one personWith backWith legsWith armsWith hard frameWith full, versus minimal or no cushioning
chair++++
sofa++/-+++
armchair++(-)+++
stool+++
bench(-)+(-)+
pew++(-)+
pouffe++

But does this overview really encapsulate the concepts you associate with the words on the left? I should have known better. For as a non-native speaker who hadn’t known the word “pouffe,” I should have concluded that this is obviously not the case. Even though I know the furniture in question, I could not imagine what the word was referring to just from looking at this table. At the same time, you can probably have a pretty good idea about what a “sofa” is, simply from lying on a sofa – you don’t need to know the other words in order to figure out the paradigmatic relations of “sofa” in order to finally have a pretty specific mental representation of “sofa” in your head. Aubrey summarizes this point as follows (p. 179):

“If the failure of theological dictionaries was the assumption that words and concepts are identical, then the failure of the structuralist semantics that dominated the field when James Barr wrote his critique was the assumption that words and concepts are dramatically different. If words mean anything at all, there must be a substantive relationship between them and the concepts (both associative and denotative) they evoke mentally.”

Was Barr wrong and Kittel right?

In other words, a dictionary entry that takes serious the insights from cognitive linguistics would actually look pretty much like an article in an encyclopaedia again, describing the “frame” of a word, as it is rooted in the social realities of the time. The hard task then is to decide which of the domains that make up the frame are evoked in any occasion. The famous example for “mother” has, for example the genetic, birth, nurturance, genealogical, and marital domains – and not all of them are usually activated together in a specific context.

All this leads us to raise the question, which must be quite shocking after what we’ve said above: So was Kittel right after all? Aubrey indeed cautiously remarks in a footnote (p. 179):

“[I]t is not entirely clear whether the theological dictionary actually confused words and concepts in the manner that they are accused or whether they were simply misunderstood by proponents of structuralist semantics. It is entirely possible that the accusation itself was a result of the great theoretical and methodological chasm that existed between historical-philological semantics and structuralist semantics.”

Where Do We Go from here?

In some sense, this is a vindication of the tradition of the TDNT, even though Aubrey also states that many authors clearly moved too quickly from psychological analysis to theological speculation in relation to words. But still, the picture just sketched might come as a relief to those who were uncomfortable with my reiteration of Barr’s criticism above. Are the suggestions above all irrelevant as a consequence? Can we just continue with our undifferentiated talk about words and concepts (and referents)? I think the contrary is the case. Sure, the just-mentioned observations point to a potential in works that explore matters of conceptuality and reference. However, most NT scholars are not well-acquainted with cognitive linguistics. (I’m certainly not.) That’s why I think we should be all the more careful when we immerse ourselves too deep in associative meanings. In any case, we should be very precise in explaining what we are talking about at any given moment. It’s fine to explore the “psychological dimension” of words as it was of interest before the raise of structuralism. But we should be very clear in indicating where we make that move. Doing that, our work will have validity even if at points we don’t capture all the relevant nuances. Others will still be able to interact with our work and build on the semantic observations that are sound.

Thus, I will close with some very practical suggestions. They don’t contain a whole lot of wisdom, mostly because I am myself an amateur when it comes to these issues of lexical semantics and lexicography. But if you are like me and from time to time in your work you make claims about the meaning of this or that Greek word, then perhaps you would also like to know some simple steps that would help a lot in increasing the transparency of your account:

  1. Read James Barr if you haven’t yet and repent of the sins of the TDNT. Then read Nida and Louw, Lexical Semantics of the Greek New TestamentThen read Mike Aubrey’s piece.
  2. Perhaps so far you have used the term “Begriff“ consistently for the ‘wesentliche Merkmale einer Sache oder einer Gruppe von Erscheinungen, die zu einer gedanklichen Einheit zusammengefasst sind’ (eWDG; emphasis mine). Well done! But other haven’t. So help us doing away with this bad habit by just no longer using this word at all. It will help a lot. Really. English-speaking colleagues will write you thank-you cards! (If you are an English-speaking scholar and that’s exactly your position, please feel free to indicate so in a comment below – it is not easy to break with bad habits and we need encouragement.)
  3. Differentiate in your terminology. First, if you are discussing a concept, say so. E.g. “das Konzept der Liebe“/“the concept of love.“ In any case, don’t use Greek words. There’s just no good reason to do so. None. We all know that you are working with Greek texts. If the concept needs specification because it is apparently so foreign to everything we know, just use adjectives. Please, in any case, don’t use transcriptions. Even if there’s a good rationale in your case, there are too many bad ones. Second, if you are discussing words, say so: “das Lexem πίστις.” (Or just the Greek word, see under #5.) To me, it doesn’t really matter whether you say “word” or “expression” or “lexeme” on the one hand and “content” or “sense” or “meaning” on the other (though they are not all the same). I am absolutely convinced that it’s better to use a slightly odd designation instead of using the entirely wrong category.
  4. To the best of your ability, avoid glosses when you try to communicate the meaning of a Greek word. Use definitions instead. That means: Louw-Nida and BDAG/Danker’s concise lexicon for those who know English. Don’t use Baur, if you are a German speaker. If you rely heavily on dictionaries, read Lee’s History. In fact, I correct myself: Do read Baur! Side by side to BDAG. It will help you recognize when their definitions are nothing but elaborations of the German glosses. Did I mention that you should read Lee’s History? (Furthermore, I know it’s a long shot but speaking of lexicons I just have to mention that it wouldn’t hurt if you learnt/deepened your knowledge of Modern Greek… there are quite a few very interesting definitions out there in Greek, if you can find and read them.)
  5. Indicate through your formatting when you are talking about concepts and when you are talking about words. In linguistics, expressions are usually adduced in italics (as in: Narrative is used for the concept ‘narrative’). That can be confusing when you use this also for emphasis. However, in philosophy and everyday usage, we tend to use double quotation marks for expressions (as in: “Narrative” is used for the concept ‘narrative’). I’d suggest that you do so when you talk about expressions in your target language. If you talk about Greek, it’s status as object language is clear anyway. You don’t have to mark it. Marking meanings typographically is a bit more difficult. In linguistics, capitals are used most often (as in: Narrative and story both mean NARRATIVE). Sometimes we even find double quotation marks, especially if the object language is different from the meta language (as in: Διήγησης“Narrative” never occurs in Paul’s letters). But that’s only a gloss anyway. I would recommend (as for example Nida/Louw do) to use single quotation marks (‘…’) whenever you talk about a concept (and you don’t actually say “concept”) or whenever you offer a definition – as in: Διήγησης can be defined as ‘discourse consisting of an orderly exposition or narration,’ which overlaps with the German expression “Erzählung,” which means ‘a text with at least two temporally joined events that are also connected in at least one further meaningful way.’
  6. Try to be as clear as possible about the “associative meanings” that you identify in a specific place and that go beyond your definition. Why do you think they are there? What does it mean for them to be “there”? Do you think it’s part of the frame of a first-century reader that would be activated in this specific context? If so, why?

Next, I would like to talk specifically about the lexical semantics of Greek verbs . Before we can do so, however, we’ll have to address the issue of aspect. We’ll tackle both in the next post!


Christoph Heilig – currently postdoc in Basel – is the author of Hidden Criticism? (Fortress, 2017) and Paul’s Triumph (Peeters, 2017). This research has recently received the Mercator Award in the Humanities and Social Sciences. Additionally, he has co-edited (with J. Thomas Hewitt and Michael F. Bird) God and the Faithfulness of Paul: A Critical Examination of the Pauline Theology of N. T. Wright (Fortress, 2017).In his most recent – and voluminous – project, which has just been completed, he discusses the importance of “stories” and “narrative substructures” for understanding Paul’s letters. It is currently in press with de Gruyter.

Was this article helpful?

Share
Written by
Tavis Bohlinger

Dr. Tavis Bohlinger is Editor-in-Chief of the Logos Academic Blog and Creative Director at Reformation Heritage Books. He holds a PhD from Durham University and writes across multiple genres, including academia, poetry, and screenwriting. He lives in Grand Rapids with his wife and three children.

View all articles
4 comments
  • I think I would now add a 7th recommendation now in light of some recent discussions: Never say that Greek word X means English/German/French etc. word Y. Words don’t mean other words. Cf. https://koine-greek.com/2020/04/22/on-literal-translation-he-that-hath-eeris-of-heerynge-heere-he/ It would be better to say that Greek word X means the same as word Y. (It will still almost never be totally right because the meanings will only overlap. But it avoids implying a category mistake.)

  • I recently read “Salvation by Allegiance Alone” where the author, Matthew Bates, proposes “allegiance” as the most accurate translation of pistis. Even though he recognizes that we should not read all the content of allegiance in every occurrence of pistis in the NT, I suspect that it is what many might be tempted to do. I wonder if, in the light of cognitive linguistics, we should be focused on frame semantics and let the context defines more accurately the meaning of the word.

    • Perhaps Logan Williams wants to say something about that? (Will message him.)
      Though I have to confess that I might also quibble over his recent comment on this on Twitter (cf. my suggestion #7):
      “Ugh, okay, ill bite: if anyone wants me to write a post arguing why pistis can’t mean “allegiance” in Paul, please strongly urge me to do so.”

  • This is good—just what I needed in thinking of some ways CL might respond to Barr, and ways Barr expresses insights CL would also hold

Written by Tavis Bohlinger
theLAB