What should a lexicon include?

Carl Conrad has raised a valid objection to calling the cumulative vocabulary list for my online grammar a “lexicon,” and I would like to address the reasons he is right. That will also give me the opportunity to lay out my long range plan for what I have called my “online lexicon.”

What is listed in the table of contents of my online grammar as the “lexicon” is not yet truly a lexicon in the correct sense of the word. This is true for at least two reasons. First, it is merely the cumulative list of the words presented in the vocabulary lists for the lessons in my grammar. It is nowhere near a complete presentation of Hellenistic Greek vocabulary. Second, what is given for each entry in the “lexicon” is not what should be included in a work that properly deserves that title. What I have included so far is nothing more than a list of English glosses for each Greek word—a list of possible ways to translate the word into English. A proper lexicon would include discussions of the meanings of each Greek word. Traditionally, Greek lexica for English speakers have also included citations of the literature upon which the lexicon is based, and I have not done this yet.

So why have I called it a “lexicon.” Well… the truth is that I have given it an ambitious title based on what I intend for it to become over time. I didn’t want to call it a “glossary” or “word list”—which might better describe its present state—because I would have to change the title later and possibly some file names, which would complicate links within the grammar. Looking back, I realize I could have made the file names conform to the “lexicon” designation while not calling it that in my blog posts, but I didn’t think of that at the time.

Here’s what I envision the “lexicon” becoming over the next few years. First, it will become a more complete list of Hellenistic vocabulary as the grammar expands. When the last chapter of the introductory grammar is uploaded, I will not stop expanding the “lexicon.” I will begin to add words not included in the grammar.

Second, I will add proper definitions—explanations of the various meanings of each word—over time, but I will not introduce these into the current document until I have completed all or virtually all of them. Writing proper lexical entries takes time and much reading of Hellenistic Greek texts. I do not want the definitions to appear piecemeal.

Third, I will add descriptions of the argument structure of all verbs, prepositions, and deverbal nouns. By “argument structure” I mean the semantic roles these words necessitate in their complements, and the relationships between those roles.

If and when that process is completed, I hope to move on to discussions of the syntactic and semantic properties of other word types.

This will likely be a life-long project. If you have input you would like to offer on particular words or on the structure of an appropriate lexicon, I would love to hear it.

I want the lexicon, eventually, to serve the interests of beginning students, readers of ancient Greek texts who need to quickly find the meaning of a particular word that has them stumped, and exegetes engaged in serious study of Hellenistic Greek vocabulary. It clearly does not meet these goals now. They are targets for the future.

Please offer feedback. I would love to hear it.

17 Replies to “What should a lexicon include?”

  1. How innovative do you want your lexicon? For example, I’ve been wondering for a while what a lexicon would look like if verbs were listed, not according to their 1st person present active indicative forms but by their aorist (or present) infinitive. Thus, instead of λύω as the head word, it would be λῦσαι. Yet, I probably will never get to see such a lexicon in my lifetime.

    1. That’s a really interesting proposal. Would you suggest the aorist infinitive or the present infinitive? The aorist infinitive would present the complication of listing as the headword a form based on a different stem from the present in the case of 2nd aorists, but that could have positive payoffs as well.

      Of course, using the infinitive is the way its done in English and many other languages.

      1. Randall Buth, in his article in Biblical Greek Language and Lexicography argued that we should be using the aorist infinitive on the basis that it is the most basic form and makes it easier for users to derive the forms for contract verbs without using a headword that doesn’t actually exist in the language (e.g. ποιέω vs. ποιῶ). He goes through all of these issues rather thoroughly.

  2. I’m glad to see that Mike Aubrey has already brought up Randall’s good paper on lemmatization. Need I say the obvious? I applaud the “lifelong” projection of this now-styled “lexicon” as the beginning of a work-in-progress (exactly what a living language is, isn’t it?). We do owe an immense debt of gratitude to Fred Danker for his own lifelong labors in lexicography (my 7-year-old granddaughter has a thing for alliteration), but it seems to me that lemmatization can and should be rationalized. Where Danker repeatedly speaks of “passives with active meaning” we should be labeling the verb as middle/passive; I think too that we should lemmatize verbs like ἐγείρομαι and ἀνίσταμαι in their middle forms and note that the actives are secondary, causative forms of such verbs. A big question to be resolved is whether to use an aorist or a present infinitive: I wonder whether this should be determined by the fundamental Aktionsart of the particular verb (obviously εἶναι and γενέσθαι would be the choices, but ἐγείρεσθαι or ἐγερθῆναι?). Oh, to have a lifetime ahead to explore and develop a brand-new Hellenistic Greek lexicon! A lexicon is the real θησαῦρος for a serious devotee of ancient Greek; it seems to me that delineation of usage, even if it is dependent on theoretical foundations and working through the requisite catagoreal choices (Aristotle said something about the importance of εὖ ἀπορῆσαι), lies at the very heart of understanding a language (“ὅπου γάρ ἐστιν ὁ θησαυρός σου, ἐκεῖ ἔσται καὶ ἡ καρδία σου “in an extended sense comes to mind!)

    1. I would like to second your comments on the contribution of Danker. While the form of his lexicon and the methodology it represents are quite traditional, his work represents a tremendous knowledge of the available Hellenistic literature and provides a wealth of examples. It would be fascinating to see the notes he took in preparation for the latest revision of the BAGD lexicon!

      I would love to hear more of what you suggest for verbs like ἐγείρομαι and ἀνίσταμαι and the secondary nature of their active forms.

  3. I know that what you are aspiring to, Micheal, isn’t quite as comprehensive as this, but I would love to see a Greek lexicon built on a similar theoretical model as the Semantic Dictionary of Biblical Hebrew (HERE)….but that might simply be because my wife is writing her thesis in the realm of cognitive linguistics…

    1. Thanks for providing the link to Reinier de Blois’ article. I didn’t know it was available online. I recommend it to anyone interested in lexicography of ancient languages. de Blois does a good job of presenting both the contributions and the limitations of time-honored approaches to lexicography as well as presenting his case for the approach used in Lowe and Nida’s Greek-English Lexicon of the New Testament: Based on Semantic Domains.

      In my own work I do want to incorporate those principles, but I think a web-ready lexicon can offer a much more usable and flexible format. Of course, I’m talking about many years into the future here. Ideally, the lexicon I have in mind should be a collaborative work, but I don’t know how I can convince others to collaborate on a work I intend to make available at no cost.

  4. Thinking outside the book: in a digital world, why choose between competing lemmatization schemes when you can support as many as make sense? A digital lexicon could be re-alphabetized on the fly with cross-reference spelling updated based on the lemmatization scheme chosen, and supporting more than one approach could facilitate interaction with a wider variety of digital media.

    1. You are absolutely right, Vincent. The digital medium offers many possibilities for extending what a lexicon can do. I need to learn a little more about database management to do all of the things I would like, but I know they can be done, including the kind of reorganization on the fly that you mention.

  5. Dear Michael,

    It appears that what you is no more than a word list that is very much similar to Metzger’s.” You note this in your response to Carl’s post. It is not a “lexicon” in the strict since of the term, e.g. BDAG. Something that is probably a closer to a cross between BDAG or Liddell-Scott would be preferable. I am almost thinking that a “lexicon”= “QHSAURUS” would be
    excellent. There are times where I see an entry and would like to see synonyms contrasted together, e.g. GINWSKW and OIDA, AGAPH and PHILH.

    Regarding the lemma used, it would be the Aorist Infinitive since I think it is the Aorist that is the most common form found in the NT (though I could be wrong on this). It would be important that certain theological terms/themes from the LXX
    be included if possible, e.g. DIKAIOSUNH, hILASTHRION, hILASMOS.

    Since there are over 10,000 verses in the NT, it would be nice for the prepositions be given more extensive detail, e.g. EN vs EIS, KATA, KAI vs DE, DE vs ALLA.

    Finally, a discussion of the words/themes used in the LXX that carry over to the NT. Just looking at USB4 with the number of quotes or verbal allusions to the OT (read LXX) is quite large.

    Like I said, “Just some thoughts.”

    En Xristwi,

    Rev. Bryant J. Williams III

    1. You raise a couple of issues I would like to acknowledge. First, the “thesaurus” idea is a great one, and this basic function could be addressed by tying each word to a semantic domain, similar to what Louw and Nida did in their print lexicon. Of course doing this in electronic form would allow for much more flexibility and the easy use of these domains to generate a thesaurus-like entry by changing a simple setting on the user interface. A similar issue has come up in the database discussion as well.

      The second issue involves what you have said about prepositions. Frankly, a lot more work needs to be done to clarify the relationship between prepositions and nominal case forms in Hellenistic Greek. Whatever is said in the lexicon about prepositions will need to take into consideration recent advances in linguistic theory and understandings of case in Ancient Greek.

  6. In parallel to the argument structure and lexical semantics you’re talking about working on, I’ve been working on comprehensive morphological information (a “morphological lexicon” if you like). Of course there’s no reason why the two couldn’t be combined and this sort of integration is what my “lemma lattice” work is designed to help with (avoiding issues with choice of lemma, at least from an integration point of view, although not necessarily for display purposes).

    In the context of the morphological lexicon I’ve been working on, I’ve been experimenting with treating voice as basically derivational morphology rather than inflectional morphology (although it’s a little more involved than that: basically I’m structuring forms not as a two level “form” mapping to a “lexeme” but instead a tree with intermediate forms, sometimes called “flexemes”).

    I’m building some tools to aid with online collaboration (see my BibleTech 2010 talk http://www.vimeo.com/10515200 ) that would be applicable not only to lexicon development but another forms of corpus-based collaboration. A prototype (focused on tasks like tagging and glossing) is available at http://oxlos.org/ with source-code at https://github.com/jtauber/oxlos )

    1. It’s great to hear from you, James.

      Thank you for the link to your BibleTech 2010 talk. It’s fascinating stuff, and I’m going to have to spend more time with it this weekend.

      I am looking for a tool to input the work I want to do in a way to make collaboration easy and uninhibited, and I’m interested in keeping as much as possible open source. It looks like you are building exactly what I’m looking for in software.

  7. After writing a brief glossary of the verbs in Galatians, I have come to the following conclusions:

    1. It is really nice and clean to provide separate entries for active verbs and middle/passive verbs. Thus, instead of ἀκούω, for example, there would also be an entry for ἀκούομαι.

    2. As much as I wanted to the use the aorist infinitive for the lemma, my previous decision forced me to distinguish middles and passives before the aorists are different for the middles and passives. This has more problems than benefits. So, I begrudgingly went back to the present. At any rate, even with the present, the aorist and other stems ought to be listed.

    3. The standard first-person present active indicative, however, does not indicate what the stem vowel of a contract verb is. The standard response to this is to use an artificial, uncontracted form, but better, in my opinion, is the present active infinitive (or possibly the third-person present active if a finite verb is desired.).

    4. If principal parts are to be listed, they should be for augmented verb forms so that its formation can be shown. This may counsel against the infinitive as the lemma.

  8. The difficulties associated with the lemma question are large. This is one reason that a digital—database driven—system would be far preferable to the print resources we have now. With the right database structure and software, the user would be able to ask for the lemma in whatever form she or he preferred, and the glossary or lexicon would be dynamically generated based on that form.

Comments are closed.