Unnecessary Simplicity in Opentext.org

An often mentioned dictum we hear from the creators of Opentext.org is, “Meaning implies Choice.”[1] This statement is not exclusive to Systemic Functional Linguistics. Rather it is foundational to much of linguistic study. Steven Runge makes a similar statement in his forthcoming Discourse Grammar, though he turns it around, “Choice implies meaning.”[2] I suggest that perhaps the maxim should go a step further, “Meaning implies structure.”[3] Applied to syntax, we find some structures that are either not represented at all or are represented poorly in the Opentext.org database.

Opentext.org assumes a completely flat clause structure so that the Subject (NP[subj]), Complement (NP[obj]) and Predicate (Verb) are all considered to be at the same syntactic level. This simplicity is helpful in explaining the apparently free order of these constituents, but it fails when we examine the distribution of adverbs.[4] With the publication of his dissertation in 1995 in the Studies in Biblical Greek series, Dr. Micheal Palmer showed rather conclusively that Greek has two different types of adverbs with their own distribution in the clause. Adverbs like εὐθέως (“instantly”) cannot occur directly following a transitive verb or an intransitive verb requiring a prepositional phrase.[5] Other adverbs have the exact opposite distribution. Πάντοτε, according to Palmer, never appears with the subject between it and the verb.[6] This is best seen in Figure 1.[7]

Fig. 1 clip_image002

The distribution of these adverbs suggests that some sort of structure prevents εὐθέως from following the verb in transitive clauses and πάντοτε from being separate from the verb by the grammatical subject. The present model of the basic sentence structure in Opentext.org cannot provide an account for this difference. In contrast, Dr. Palmer’s model does explain the difference in distribution. Πάντοτε would be considered a Verb Phrase level adverb, while εὐθέως would be viewed as a Sentence level adverb. If it is true the meaning implies structure, Opentext.org’s present analysis prevents us from being able to discover the semantic difference between such clause level adverbs and non-clause level adverbs.

But at the very same time, Palmer’s analysis cannot be accepted either. Because he includes a Verb Phrase in the Greek clause structure, he cannot explain the distribution of the subject constituent in ditransitive clauses, such as in Acts 16:38 in Figure 2 below.

Fig. 2 clip_image004

In this clause, the subject appears between the indirect object and the direct object. Typically, if a language has a Verb Phrase, both objects would appear inside. But Acts 16:38 suggests that the subject would be in the VP as well.[8]

The point of all of this is not to say that Greek has a Verb Phrase, but to show that Opentext.org’s analysis, in its caution, cannot explain certain syntactic phenomena. Regardless of what one concludes regarding πάντοτε, it is quite clear that the adverb εὐθέως does have a unique syntactic distribution that should be represented in a syntactic analysis. An acceptable analysis of Greek syntax will need to accurately represent both the distribution of different types of adverbs and also the significant variations possible in the order of subjects and objects.

[1] Stanley E. Porter, Verbal Aspect in the Greek of the New Testament with Reference to Tense and Mood (SBG 1; New York: Peter Lang, 1989), 12; cf. Jeffery T. Reed’s statement, “[SFL] is a theory of language as choice” (A Discourse Analysis of Philippians: Method and Rhetoric in the Debate over Literary Integrity [JSNTS 136; Sheffield: Sheffield Academic, 1997], 36. It should be noted that this sort of statement is not limited to Systemics.

[2] Steven Runge, Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis (LBRS; Bellingham, Wash.: Logos Research Systems, forthcoming), section 1.1.

[3] Stephen David Ross, The Limits of Language, (New York: Fordham University Press, 1993), 34.

[4] Distribution refers to the places in a sentence of phrase where a given part of speech can appear and still be grammatically correct.

[5] Micheal W. Palmer, Levels of Constituent Structure in the New Testament Greek, (SBG; New York: Peter Lang, 1995).

[6] While Palmer’s claim regarding the distribution of εὐθέως holds true across the corpus of the GNT, Apostolic Fathers, Philo, and Josephus, his claim about πάντοτε does not have as firm of a basis. On one hand, across the same corpus of texts πάντοτε and the verb are never separated by a noun in the nominative case (with the exception of an error in Lightfoot’s edition of the Apostolic Fathers at Herm, Man. III, 1, “πάντοτε πανοῦργος ἔζησα”). But on the other hand, there are in fact only eight instances in the NT where πάντοτε occurs in a clause that has an overt subject. Likewise, there are no instances in the Apostolic Fathers where πάντοτε appears with an NP Subject. The same holds true for Philo, where πάντοτε only occurs twice at all. In Josephus, the adverb occurs four times and only once does it occur in a clause with a subject (Antiquities 15.140) and in that instance, it is a verbless, non-verbal predicate, which by definition cannot have a Verb Phrase. Finally, in the significantly larger corpus available through the Perseus Digital Library, πάντοτε appears 71 times and in the occurrence not already mentioned (25), none of them occur in clauses with an overt subject. This leaves me suspicious of Palmer’s claim for this particular adverb, though its possible that he has found other adverbs with this distribution and that he has access to the TLG where he could analyze a larger number of occurrences.

[7] Adapted from Palmer, Constituent Structure, 45.

[8] The theory in which Dr. Palmer works would assume a movement rule that scrambles the phrase structure so that the subject appears within the Verb Phrase, but it must be questioned what sort of evidence there is in Greek for such a thing, particularly when the general consensus that the default constituent order is Verb initial.