Disclaimer: The post below refers the implementation of SFL in NT studies, which may or may not necessarily reflect Halliday’s intentions. My goal is to address the methodological and theoretical flaws of the approach to SFL popularized by Stanley Porter. There is reason to believe that it represents a departure from Hallidayan linguistics.

My focus in Greek for the next while will be developing a description of the historical present usage observed in the Greek NT, with special attention to how this usage informs current descriptions of verbal aspect. Most notably, these will include the work of Porter, Decker and Campbell. All three (and I believe Fanning as well) utilize a systemic-functional framework for their description. For this reason, it behooves me to devote attention to the framework itself before attending to their finished product. I contend that several core elements of systemic functional linguistics limits its utility as a descriptive framework for the complexities of the verbal and syntactic[1] systems. In order to conduct this review, I will draw upon the explanations provided by those applying SFL to study of Koine Greek.

Systemic Functional Linguistics (SFL) usefully conceives of language as a system of options or choices. The overall system of a language is composed of smaller sub-systems, which in turn have smaller sub-systems, with the individual parts building recursively into the whole. Reed uses the system of grammatical person, which is a sub-system of the larger referential system, to illustrate.

person_tree

When one is parsing the morphology of a pronoun, there are three different options available regarding person, determined by the semantics of the context. One of the unfortunate terms that SFL uses is the word “choice” for what is probably better called an option. In the case of person, I cannot simply choose the form I want, and make up the meaning to go with it. Indeed, semantic factors play an overarching role in the decision making process.

Note also that these options are expressed in the form of a tree. Trees work very nicely for simple, univariable decisions, as with person. SFL is well suited to morphology, since each factor of case, number, gender, person, etc., can essentially be considered on its own, i.e. in isolation from the other discrete factors, without losing much in the process. Trees provide a tidy mechanism for displaying options in a series, where they do not simultaneously interact with one another.

Below is another example of a tree, taken from Decker’s Temporal Deixis of the Greek Verb in the Gospel of Mark with Reference to Verbal Aspect, p. 107 describing aspect and remoteness:[2]

aspect_remotenessThis diagram does a nice job of illustrating the various constraints that are unique to each verb form. Note that person and number are omitted since they are common to all forms. The purpose of this tree is to show how the aorist differs from the present, the present from the perfect, and so on.

One of the primary purposes of linguistics is to describe language usage. You may say “Duh”, (or if you are more cultured, possibly “d’accord”). Note that the trees utilized in SFL are limited to considering only one factor at a time. In this case, the tree considers perfectivity. Based on the answer to this constraint, one moves on to the kind of non-perfective action: imperfective or stative. Finally, one moves on to consider remoteness. Each different constraint is considered on its own in isolation from the others. The ordering of the decisions represented in the tree is based upon simplification of the decisions, not necessarily the order that speakers would make them. For instance, both stative and imperfective have +remote and -remote. In theory, the tree could have moved from -perfective to consideration of remoteness rather than imperfective/stative. Doing so would have left two forms (e.g. present and imperfect) with the same characteristic (+imperfective) on separate branches. Organization into like kinds drives the organization of the tree more so than the writer’s/reader’s actual decision-making process.

WARNING: Lingusitic principle ahead! Linguistic descriptions, especially functional ones, are to model the actual usage as closely as possible. I do not think that Decker intends his figure to represent the actual order of decision making in choosing a verb form. Having said that, the discussion of the usage of various verb forms by Porter and others nonetheless seems to presuppose such a tree.

Porter describes verbal aspect as “a semantic (meaning) category by which a speaker or writer grammaticalizes (i.e. represents a meaning by choice of a word-form) a perspective on an action by the selection of a particular tense-form in the verbal system.[3] The discussions of choice and usage seem to mimic the tree, in that one consideration at a time is considered. The use of trees by SFL leaves little choice, since it is not designed to represent multi-variable matrices.

Trees can be a very useful way of describing available options, allowing one to focus discretely at one variable at a time. Unfortunately, the use of trees and attending to only one variable at a time has influenced the description of the Greek verbal system. It began with Porter, and others have built upon his foundation. Here is another example of a tree-based description from Porter and O’Donnell’s article in Filología Neotestamentaria – Vol. XIV – 2001, pp. 3-41:

Again, this is a tidy and impressive way to describe the options that are available, but it begs the question of how closely it models the actual decision making of the speaker. It the modeling is off base, the tree becomes a theoretical exercise rather than a functional description.

There are numerous linguistic methodologies out there in the wider world beyond NT studies, but you would hardly know it from discussion within. Literature surveys by practitioners of SFL are horrible oversimplifications, if not just wrong at points.((See Reed’s  Generally speaking, different methodologies were developed to address problems and data left unaddressed by previous methodologies. SFL is well-suited to describing simple systems of language, ones that do not have multiple variables simultaneously in operation. Halliday primarily designed SFL for describing language features of English, which is rather simple in both in terms of syntax and the verbal system. Our verbs are built up using discrete building blocks, such as helping verbs, to achieve various forms. Koine does not break down so simply. English syntax is constrained severely in comparison to Greek.

As I said earlier, SFL is well suited to describing rather simple, univariable systems. It stumbles when life gets more complex. I find it interested that nearly all descriptions of the Greek verbal system utilize SFL. I think that SFL has taken us about as far forward as it can, based on its inherent limitations. I will be using a different model for my description of the historical present, one that is designed for complex matrices.

My master’s thesis adviser told me to pay attention to my presuppositions, being sure to disclose them to the reader. When I started looking at the work of others, I found that few people do this. In fact, presuppositions predispose you to follow certain paths. This can be a good thing or a bad thing, depending upon the situation. I contend that the presuppositions of the SFL framework predispose the analyst down certain paths, as illustrated by the tree. Even though none would probably claim that Decker’s tree models the decisions made by a speaker regarding verbal aspect, it inevitably affects the conception and discussion of the decision-making process. The SFL framework has forced people to paint themselves into certain corners. If one conceives of the world as trees or clines, everything must fit into one of the two categories, whether it fits well or not. I will return to this issue in my next post, looking at the area of markedness, and SFL’s use of clines to create hierarchies.

In what ways do you think that the use of trees has influenced the conceptualization of the Greek verbal system? I would like to hear your thoughts on the matter.

[1] For a comparison of approaches to describing information structuring in languages, see Runge, Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis, §9.4.

[2] I recreated the tree in Word rather poorly, so this is the content, not the exact likeness, of Decker’s figure.

<[3]Stanley E. Porter, Idioms of the Greek New Testament (Sheffield: JSOT, 1999), 21.