This continues a series on Markedness and Verbal Aspect devoted to comparing Porter’s claims derived from the linguistic literature against what the linguists themselves say about the matter. I only rely upon the sources he cites, no new data is introduced. I have already demonstrated that he fundamentally misunderstands the grounding literature (i.e. foreground/background, see posts beginning here), deriving conclusions that lack any basis in his cited sources .

You might ask, “Who cares? Why does it matter?” Porter’s  entire theoretical framework about the prominence of the Greek verbal aspects relies upon four supposedly different kinds of markedness: material, implicational, distributional and semantic .1 If his understanding of these concepts is wrong, then his entire theoretical framework is in jeopardy, a house of cards. Back in May I discussed the first two types of markedness (here) , and will tackle semantic markedness is this present post.

Porter claims the work of Zwicky and Comrie as his authorities on markedness.2 Porter never really defines what he means by semantic markedness, but makes the following statement in regard to what can be gleaned from it:

In chapt. 2, four different analogies are drawn–verbal opposition, conceptual description, planes of discourse, systemic framework–to describe the semantic markedness of the individual verb forms. Through each one it can be seen that the Aorist is the less heavily marked verb form in the Present/Imperfect opposition, and the least heavily marked when the perfect is also considered.3

Note that he describes semantic markedness quantitatively, as though each aspect has some numerical value derived from the factors he discusses. Porter’s quantitative understanding is unique in that the linguists he cites understand markedness to be qualitative in nature. He understands them to be talking about the amount of marking that is present, inferring without any support from the literature that the amount of markedness present is directly proportional to the form’s prominence.

It is important to note that  Zwicky and Comrie are conducting typological studies to find the most basic forms in language compared to the other members of the set. They are interested in the presence or absence of markers, not in totaling them up to establish one form as more semantically marked than another. See the posts covering this idea here and here.

Zwicky and Comrie appear to be talking about something fundamentally different than Porter in their discussion of semantic markedness. We begin with Comrie’s treatment, since it informs Zwicky’s ideas to some extent.

Comrie explains his interest in markedness for the study of verbal aspect by noting:

The intuition behind the notion of markedness in linguistics is that, where we have an opposition with two or more members (e.g. perfective versus imperfective), it is often the case that one member of the opposition is felt to be more usual, more normal, less specific than the other (in markedness terminology, it is unmarked, the others are marked).4

After noting that one cannot rely on intuition alone for such decisions, Comrie proposes using the following criteria: semantic, morphological, and statistical markedness, describing the first one in this way: “One of the most decisive criteria is that, in many cases, the meaning of the unmarked category is always optional, i.e. where the unmarked category can always be used, even in a situation where the marked category would also be appropriate.”5 He cites Italian and Spanish as representative examples wherein the progressive/imperfective forms have a similar meaning to English, “however in Spanish and Italian these forms can always, without excluding progressive meaning, be replaced by the non-Progressive forms scrivo, escribo, whereas in English changing I am writing to I write necessarily involves a shift to nonprogressive meaning.”6 He notes the same is often true in most Slavonic languages, but not identical.

What does Comrie mean by semantic markedness? He is referring to instances where one member of a set may be used e.g. for either perfective or impefective action. It is only with the use of the marked member that the exact meaning is disambiguated. The unmarked member lacks a marker to signal which meaning is intended. The marked form, therefore,  has a more specific meaning.

Zwicky, Porter’s primary source for markedness, devotes only eight sentences to semantic markedness in “On Markedness in Morphology”7 He uses nouns as examples, noting the various options for referring to the general concept of horse in English. Mare, stallion, and filly are each more specific than horse, but horse may appropriately be used to refer to any one of these three. Horse is unmarked for age or gender, whereas mare and stallion are marked for gender (female and male, respectively); whereas filly is marked for both gender and age (i.e. a young female horse).

When these linguists talk about semantic markedness, they are describing a general/specific relation that exists in the language. They are referring to an unmarked form that can be used for a general reference, whereas another, more specific form (the marked one) may also be used to disambiguate which specific thing is intended. The unmarked progressive in Spanish and Italian may have either a progressive or non-progressive meaning. There is no marker present to specify which is intended. Based on the congruity of the examples, they are clearly talking about the same concept. The same cannot be said of Porter’s conception. Although Porter cites Zwicky and Comrie as primary sources for markedness, his conception of  semantic markedness bears no resemblance to their discussion of the matter.

Here is where it gets interesting. Porter has claimed that the aorist tense-form is the least semantically marked form in Greek. According to Comrie and Zwicky’s conception of semantic markedness, we would expect that a perfective form could be used with some other meaning than perfective being possible, i.e. a stative or imperfective meaning. Furthermore, we would expect that an imperfective or stative form could appropriately be substituted in the same context without fundamentally changing the meaning; it should match one of the possible meanings when a perfective was used. In other words, the aorist form should include the other meanings of the marked forms. However, Comrie himself points out that this is not the case in Ancient Greek:

The application of this particular criteria to many other aspectual oppositions such as Progressive/non-Progressive in English or the Simple Past/Imperfect in Spanish, or the Aorist/Imperfect distinction in Ancient Greek (and those Slavonic languages that retain these forms) is more problematic, since the usual pattern here is for the categories to be mutually exclusive, i.e. the replacement of the Aorist by an Imperfect or vice versa usually implies a different meaning altogether, not merely loss of some information by use of an unmarked category. It is generally, though not universally, assumed that in Romance and Slavonic languages, and in Georgian, it is the Aorist (Simple Past) that is the unmarked member of the Aorist/Imperfect opposition, but the possibility cannot be excluded that we are here dealing with two equally marked members of an equipollent opposition. 8

So what does this mean? To begin with, Porter must be using semantic markedness in some other sense than Comrie or Zwicky, yet without stating so. If he is using the same meaning, it should have been clear from Comrie’s work that any claim of semantic markedness in Greek would be invalid, based on the change of meaning that comes about with the change in aspect. Greek does not manifest the same unmarked freedom of usage described in the examples from Spanish and Italian aspect or from the English horse.

Comrie’s note about Greek’s equipollent marking of aspect (i.e., that each form distinctly marks a different aspect, as opposed to one being unmarked like Spanish or Italian) was not lost on Porter, as he actually makes the very same claim! Porter states,

There is no apparent evidence that in Greek any of the verbal aspects is semantically unmarked (contra Haberland, “Note,” 182). In fact, this work argues that even within the binary oppositions all members contribute semantic weight to the verbal component of the clause. Greek verbal aspect, therefore, appears to function on the basis of equipollent binary oppositions, in which each aspect is not identically weighted, at the least each contributes semantically in an identifiable way (see Fredrich, “Theory,” S14) [sic].9

If this claim is true regarding equipollency–and I would agree that it is–there is absolutely no basis for Porter to make claims about semantic markedness of the kind described by Comrie and Zwicky. If he intends to make it on some other grounds, where is the justification for it?

At the end of this quote, Porter mentions the notion of “semantic weight.” This is yet another theoretical concept that appears to originate from Porter to justify a quantitative understanding of these criteria (see here for more on this matter). The equipollent marking means that each form uniquely signals the presence of a distinct set of non-overlapping semantic features. According to his own literature used to support his discussion, there are no legitimate grounds for claiming semantic markedness as a valid criteria for Koine Greek.

I am not really sure how to conclude this post, other than to say that Porter’s use of the linguistic literature bizarre; semantic markedness is yet another of the four theoretical pillars on which he argues  for his theory of aspectual prominence, yet the literature he cites for support actually argues against it. For semantic markedness to be a valid criteria for use in Greek, Porter would need to argue that at least in the case of the Aorist the aspectual system is not equipollent.

I repeat the assertion with which I began this series:

There was a curious comment from Moises Silva that will function as my point of departure. He states:

I suppose that this problem lies behind a curious thing. In general terms, I found Porter’s theoretical framework more convincing than Fanning’s.1 Fanning is, I think, much too generous in his attempt to salvage what he can out of the traditional grammars; while I commend him in the attempt (it needs to be done), the result is a certain instability. On the other hand, when it came to looking at their implementation of the principles, I had many more problems with Porter than with Fanning: time and time again I failed to see either the logic or the evidence for his interpretations. I am not ready to suggest that we adopt Porter’s explanation of the linguistic system and Fanning’s interpretation of actual occurrences. My point is only that some important inconsistencies remain in the field as a whole, and that we all need to clarify more precisely what we are trying to do.2

At the risk of incurring the boundless wrath of one Esteban Vasquez, I will merely add to the “Infallible One’s” point (with gentleness and respect no less) rather than disagree with it. Here is the key thing to recognize: if there are problems with the implementation, then there are likely problems in the theoretical framework that guided the implementation.

I think Silva was onto something far bigger than he may have known at the time. If you have problems or doubts regarding the claims I am making here, read Porter’s primary sources for yourself. Too many NT scholars have uncritically accepted his claims without testing them against the linguistic literature. You need not go any further than Porter’s own sources to find evidence against his claims. His basic definitions and classifications of the Greek tense-forms are sound, I have affirm these things myself. But his claims regarding their backgrounding/foregrounding/frontgrounding function, regarding their prominence value on the basis of markedness, are discredited by the very sources he cites in support of them.

One more post remains to finish this series before I submit the findings for publication and present them at ETS in Atlanta.

Return to On Porter, Prominence and Aspect.

  1. See Verbal Aspect in the Greek of the New Testament: With Reference to Tense and Mood (Studies in Biblical Greek 1. New York: Peter Lang, 1989), 179-181. [↩]
  2. Ibid., 89. [↩]
  3. Ibid., 181. [↩]
  4. Bernard Comrie, .Aspect: An Introduction to the Study of Verbal Aspect and Related Problems (Cambridge Textbooks in Linguistics, Cambridge; New York: Cambridge University Press, 1981), 111. [↩]
  5. Ibid., 112. [↩]
  6. Ibid. [↩]
  7. Arnold M. Zwicky, “On Markedness in Morphology,” Die Sprache (Zeitschrift für Sprachwissenschaft Wien 24, no. 2 , 1978), 130-31. [↩]
  8. Comrie, 113-114. [↩]
  9. Porter, Verbal Aspect, 90. [↩]