Telicity of ἐλθεῖν (ἕρχομαι)?

[MAubrey]

(Sorry about that, but I couldn't help it.)

It's because you can't help it, Carl, that I like you so much.

[MAubrey]

Espinosa talks about ἐλθεῖν being "lexically telic" and "inherently deictic" but are those the same as or different from your "inherently telic"?

Well, he bases his 'lexical telicity' on the basis that the verb 'denotes motion toward the deictic center', so they seem the same to me. Either way, I'm not sure that's a very good inference for him to make.

And what's your objection to it? The general concept? Or, as as applied to ἐλθεῖν?

Sorry...so much discussion happening in so many separate posts...

As my wife put it while we were discussing this issue before going to sleep last night, all that Espinosa has really done is described the nature of entities moving through space. An implicit deictic center doesn't then necessitate telicity.

The fast answer to the question in terms of how I understand this verb is this:

ἐλθεῖν's prototypical instantiations are certainly telic. However, telicity is not a schematic feature of its larger usage...just like flight is a necessarily feature for the prototypical bird. It's required for the prototype, but is not obligatory for the schema as a whole: penguins and Emus would be very disappointed if it was. There's always a prototype and there's always a schema. And they rarely (and by rarely, I probably mean "never") coincide.

[Stephen Carlson]

Sorry...so much discussion happening in so many separate posts...

Thanks for taking the trouble to respond. I'll try to bring together the different scattered responses.

The fast answer to the question in terms of how I understand this verb is this:

ἐλθεῖν's prototypical instantiations are certainly telic. However, telicity is not a schematic feature of its larger usage...just like flight is a necessarily feature for the prototypical bird. It's required for the prototype, but is not obligatory for the schema as a whole: penguins and Emus would be very disappointed if it was. There's always a prototype and there's always a schema. And they rarely (and by rarely, I probably mean "never") coincide.

Well, I have to admit that when you said that Espinosa's claim was " incontrovertibly false", I didn't expect that this judgment would mean something along the lines of, while ἐλθεῖν is prototypically telic, there a few cases where it is not (some of which were outside of the corpus he looked at). It was particularly surprising to me because the context Espinosa was mentioned in was a comparison of his thesis with that of Shain, who concluded that ἐλθεῖν denoted an activity. Given this context, it sounded like you were implicitly agreeing with Shain over Espinosa. While neither of them used a prototype approach, it was striking that the one who came closer in characterizing the prototypical instantiations as telic was the one singled out as "incontrovertibly false." Well, okay.

Now my own sense, in working with this verb, is that ἐλθεῖν is a fairly good example of verb whose situation type is an achievement (i.e., telic), except for possibly some fringe cases. That's why I felt that Espinosa had the better of the argument over Shain. Don't get me wrong, I do have a number of issues with Espinosa's thesis. Iv'e already pointed out the problem with 'toward" instead of "to" a destination. I could probably add that he misses the mark with proposing the (a?) deictic center as the destination as well.

As my wife put it while we were discussing this issue before going to sleep last night, all that Espinosa has really done is described the nature of entities moving through space. An implicit deictic center doesn't then necessitate telicity.

I think he's doing more that describing entities merely moving through space. He asserts that moving to a destination (τέλος!) is part of the denotation, and that's what makes it telic. He also tries to identify this destination as the/a deictic center, but that is less persuasive to me (though it may be true for some or even prototypical instances of ἐλθεῖν), which we could (or could not) get into.

But clauses like: δόξα θεοῦ Ισραηλ ἤρχετο κατὰ τὴν ὁδὸν (Ezekiel 43:2) or ἦλθεν γὰρ Ισραηλ ὁδὸν Αθαριν (Numbers 21:1) tell me that the verb isn't telic. I also think his analysis of Matt 15:29 is getting awfully close to question begging.

I admit I was initially baffled by your first example, since ἤρχετο is imperfect and imperfectives of telic situation types do not entail that the end point is reached. After some thought in how to reconstruct the implied premise of your argument, I'm guessing the point is perhaps that these examples all have path adjuncts (κατὰ τὴν ὁδόν, ὁδὸν Αθαριν, and maybe παρὰ τῆν θάλασσαν), which are presumably more suitable for atelic activities than telic achievements. Is that the argument behind these examples? Or is there some other reason that these example tell you that the verb (presumably in these examples) isn't telic? At any rate, it is not clear that any of these examples really exclude the position that there is an implicit destination lurking inside of έλθεῖν.

[MAubrey]

Well, I have to admit that when you said that Espinosa's claim was " incontrovertibly false", I didn't expect that this judgment would mean something along the lines of, while ἐλθεῖν is prototypically telic, there a few cases where it is not (some of which were outside of the corpus he looked at). It was particularly surprising to me because the context Espinosa was mentioned in was a comparison of his thesis with that of Shain, who concluded that ἐλθεῖν denoted an activity. Given this context, it sounded like you were implicitly agreeing with Shain over Espinosa. While neither of them used a prototype approach, it was striking that the one who came closer in characterizing the prototypical instantiations as telic was the one singled out as "incontrovertibly false." Well, okay.

Let me see if I can take a step back and explain where I'm coming from. I was implicitly agreeing with Shain over Espinosa. Absolutely (though that isn't to say that I do agree with here...but that's a whole other can of worms. The issue here is that their sort of binary +/- approach to semantics isn't, as you rightly noted, prototype semantics. So the question of what is and what isn't the prototype isn't a relevant question here at all--bringing in prototype theory in my previous post was intended solely as a summary of my view of the nature of this verb and its usage independent of either Shain or Espinosa. What they're talking about can be construed as talking about the 'basic meaning' meaning of the lexeme. The prototype isn't in play. All they care about is the general meaning of the lexeme. When we start talking about general meaning in lexical semantics, we're now talking about the overarching schematic structure of the lexeme's usage. Coming back to my bird example, once we're talking about the overarching schema, it doesn't matter whether the prototype is a dodo or a bald eagle. The defining characteristics of that prototype are no longer defining characteristics for the category as a whole. So coming back to the verb in question, to say that ἔρχομαι is "lexically telic" is to make a claim about the defining characteristics of the category: i.e. the schema. And to that end, I think Espinosa has missed the mark here.

Am I making any more sense than before...?

Now my own sense, in working with this verb, is that ἐλθεῖν is a fairly good example of verb whose situation type is an achievement (i.e., telic), except for possibly some fringe cases. That's why I felt that Espinosa had the better of the argument over Shain. Don't get me wrong, I do have a number of issues with Espinosa's thesis. Iv'e already pointed out the problem with 'toward" instead of "to" a destination. I could probably add that he misses the mark with proposing the (a?) deictic center as the destination as well.

Whether Espinosa has a better argument or not depends on the question at hand. For me, not all lexemes inherently have a situation type. Some certainly do, but many (most?) do not (this view is an attempt to mediate conflicting claims between those who hold that situation aspect is totally lexical and those who hold that situation aspect is totally contextual/usage-based. I do accept that lexeme can involve particular conceptual categories, like telicity.

Now, while binary features are cognitive lies about the nature of categories, for practical purposes they are extremely useful cognitive lies, a convenient fiction, you might say.

[+/−static]
[+/−dynamic]
[+/−telic]
[+/−punctual]
[+/−causative]

You'll recognize these as the features that are used to define situation types. I add causative because it makes for a much cleaner categorization and also brings situation aspect in line with larger discussions of verbal semantics (talking about situation aspect without respect to causation is like trying to talk about middle voice without subject affectedness). That is, in nearly all aspectual theories, situation types aren't real. They're the total sum of these features (though again: cognitive lies). So lexemes don't have classes. They have conceptual features that are then used to define the classes. And that's what I see when I look at English verbs:

eat
go
summit
believe
shatter

Some of these verbs inherently denote a sufficient number of features in their lexical semantics to actually convey a situation type without a clause. Believe can only be a state since [+static] is all that's require for a state to be a state. Likewise, summit can only be an achievement because it's [+dynamic], [+telic], and, [+punctual]. All three of those are necessary conditions for achievements. What about shatter? Definitely [+punctual] and definitely [+dynamic]. But is it transitive [+causative] or intransitive [−causative]? These are separate senses and there is no clear boundary as to whether one is more primary than the other. The sense is determined by the predication in question and in turn determine the situation aspect.

The window shattered.
John shattered the window.

And what about eat? It is certainly [+dynamic]. And it probably has duration, so it's also [−punctual]. But is it telic? There really is no way of knowing without it appearing in complete full predication:

John ate peanuts all morning. [−telic]
John ate bacon and eggs for breakfast. [+telic]

Eat, as a lexeme, is unmarked for [+/−telic]. The English go is similar. Again definitely [+dynamic], but is it [−punctual] or [+punctual] and is it [+telic] or [−telic]? These two a tied together depending on the clause:

John went home on his bike yesterday.
!!John went home on his bike yesterday for an hour.

When go has an explicit endpoint [+telic], it is definitely [+punctual]. In such clauses it cannot take a duration PP. In the sentence above, for an hour can only be construed as referring to how long he stand. It cannot be construed to referring to how long his coming took. As such, in these clauses go must be an achievement. However...

John went along the lake on his bike yesterday.
John went along the lake on his bike yesterday for an hour.

Go doesn't need an explicit endpoint to be grammatical. It isn't lexically telic. It is unmarked for [+/−telic]. And in turn, it then also allows for duration as well. And this is true regardless of whether viewpoint aspect is imperfective or perfective.

All that to say is that lexemes don't have situation classes. They have the building blocks for them, but they do not have them. In that sense, Espinosa is asking the right question, but he's coming to the wrong conclusion.

As my wife put it while we were discussing this issue before going to sleep last night, all that Espinosa has really done is described the nature of entities moving through space. An implicit deictic center doesn't then necessitate telicity.

I think he's doing more that describing entities merely moving through space. He asserts that moving to a destination (τέλος!) is part of the denotation, and that's what makes it telic. He also tries to identify this destination as the/a deictic center, but that is less persuasive to me (though it may be true for some or even prototypical instances of ἐλθεῖν), which we could (or could not) get into.

There's a lot that I could say here, but this post is already really long and it would lead away from the main discussion into peripheral questions about the nature of motion verbs through space. We can save that for another time, perhaps.

But clauses like: δόξα θεοῦ Ισραηλ ἤρχετο κατὰ τὴν ὁδὸν (Ezekiel 43:2) or ἦλθεν γὰρ Ισραηλ ὁδὸν Αθαριν (Numbers 21:1) tell me that the verb isn't telic. I also think his analysis of Matt 15:29 is getting awfully close to question begging.

I admit I was initially baffled by your first example, since ἤρχετο is imperfect and imperfectives of telic situation types do not entail that the end point is reached.

The imperfective example is less baffling if you consider how the imperfective aspect gets (or does get) used with verbs that truly are lexically telic. Take the verb summit again, which is truly lexically telic.

Nobody would say: *John was summitting the mountain yesterday, but he didn't make it to the top. That's semantically illformed. The imperfective aspect for this verb generally work best only in a list of events, where the imperfective says nothing about the individual event, but plays a larger discourse-pragmatic role in terms of a larger progression:

For his big campying trip, John was summitting the mountain on Thursday and then going fly fishing on Friday and Saturday.

Or the imperfective could be used when the speaker has a gap their knowledge:

A: Hey have you seen John? I was supposed to meet him for coffee this afternoon.
B: No, I have. He was summitting the mountain yesterday...Oh dear. I hope he's okay.

Independent of these situations, to be useable, the incompleteness needs to be primed in the mind of the audience before the introduction of the lexically telic verb. This is usually done by means of subordination. Thus, the imperfective aspect is already expected by the matrix verb in a clause like:

John was supposed to be summitting the mountain today.

For the imperfective to work in this case, the situation must be explicitly presented as being contrary to fact. And actually, this sort of construction would probably be preferred to the bare imperfective in the A/B dialogue above. Anything with going to be, supposed to be, was intending to be, planned on, etc. Priming imperfectivity is going to be preferred.

The point is that in all these instances: in progress lists, information gaps, and contextual priming for cancelability, the larger context and usage plays a significant role in communicating a reason why a given lexically telic verb is being used in the imperfective aspect. None of that is happening in the imperfective Ezekiel example I provided.

[Stephen Carlson]

Thanks for your response, Mike. It's pretty long and I appreciate your writing. I would like to focus on the Ezekiel example and what it says for the telicity of ἐλθεῖν, but I would like to first register my agreement or at least my non-disagreement with the idea that some lexemes are neither telic or atelic by themselves. Eat is a great example; it's not really clear whether its inherent situation type is an accomplishment or activity.

There are a number of theoretical options for dealing with this that I've seen. For example, one can say that eat is neither telic nor atelic, and it depends on its arguments, etc. (I think that's your position). Or one can simply choose one of them as the basic situation type and hold that the other situation type arises from some of kind of aspectual coercion of the basic situation type when triggered by an appropriate argument etc. Or, one can say it's both, i.e., that eat is polysemous and the contextual clues and arguments select one of the types for the context.

Whichever approach one takes, it's irrelevant to issue at hand, because I don't think έλθεῖν is one of these kind of verbs. Rather, I think it falls in a class of achievements that Fanning calls a climax and Croft calls a run-up achievement, and others a process achievement. (Note that not every aspectologist recognizes this class: Carlota Smith thinks they are simply accomplishments (which are still telic), and Huovila suggests that they are really polysemous between achievements and accomplishments, both of which are telic.) These are achievements in that they do have a built-in end point (for ἐλθεῖν it would be the destination) and no minimal extent and no built-in start point. Unlike other achievements, they are associated with a prefaced process. Examples of such verbs in English include reach, grow up, and die.

This type of verb, though telic, has the behavior that when in the imperfective, they have a reading that Comrie calls "prospective," but I've also seen called point progressive, boundary progressive, imminential, propinquitive, etc. The label doesn't matter as long as the concept is understood. This reading is that the culmination of the event is imminent, or about to happen, or on the verge of happening. When I look at Ezek 43:2, that's what I see:

καὶ ἰδοὺ δόξα θεοῦ Ισραηλ ἤρχετο κατὰ τὴν ὁδὸν τῆς πύλης τῆς βλεπούσης πρὸς ἀνατολάς, καὶ φωνὴ τῆς παρεμβολῆς ὡς φωνὴ διπλασιαζόντων πολλῶν, καὶ ἡ γῆ ἐξέλαμπεν ὡς φέγγος ἀπὸ τῆς δόξης κυκλόθεν.

Here, the glory of the God of Israel is on the verge of coming, Ezekiel can hear it and he see the earth shine around in reaction. In v. 4, this glory of the Lord actually comes and enters the temple. If έλθεῖν is such an achievement, this reading is a perfectly fine example of it and I don't see any need to upend this rather standard and boring treatment of this type of verb.

The point is that in all these instances: in progress lists, information gaps, and contextual priming for cancelability, the larger context and usage plays a significant role in communicating a reason why a given lexically telic verb is being used in the imperfective aspect. None of that is happening in the imperfective Ezekiel example I provided.

So I did guess wrong about why you cited Ezek 43:2! As for your argument... well, I've read a boatload of literature on aspect and I haven't seen anyone impose such restrictions on the felicity of the imperfective with such telic verbs. That doesn't necessarily make it wrong, but it does make it harder to evaluate. Is this your personal theory or is it described somewhere in the literature? This is not an idle request: I'd love to understand it more, and I'm happy to read up on pertinent literature you can point out, but right now it just looks like something from another universe to me.

(It also doesn't help that the verb used to illustrate your examples, "to summit," is not part of my vocabulary. It's not even listed in the OED, though it lists other senses for "to summit" (e.g, the obsolete "to submit" or the cold-war "to take part in summit meetings"). It appears to be hikers' jargon, which I don't have any feel for: even if I try glossing "to summit" as "to reach the summit," some of your examples sound kind of odd.)

[MAubrey]

Sorry I haven't replied yet, Stephen.

I'm in a rush before SBL to get a sort of preface up on my blog before I make my thesis publically available and that's taking up a lot of my free time. This thread is book marked and I'm going to try to get back to this soon. Worst case scenario would be that could answer your questions in person at SBL.

[Scott Lawson]

Mike et al,
I'm just a dummy but I have found this discussion quite fascinating and would hope to see your response to Dr. Carlson. And to Dr. Carlson your objection to "summit" as a verb seems a bit disengenous since it could easily be substituted with the word "crest" which is recognized as a verb. It felt to me like a "gotcha" rather than a genuine engagement of his points. I'm sorry if I am overstepping my bounds in so stating this. I respect both of you deeply as scholars and enjoy learning from you both but Stephen your last response to Mike seemed to be more of a conversation ender than designed to continue the dialog. My apologies if I am mistaken in my impressions. I hope to see a continuation of this topic if possible by you two.

My best regards,

Scott Lawson

[Stephen Carlson]

Sorry I haven't replied yet, Stephen.

I'm in a rush before SBL to get a sort of preface up on my blog before I make my thesis publically available and that's taking up a lot of my free time. This thread is book marked and I'm going to try to get back to this soon. Worst case scenario would be that could answer your questions in person at SBL.

I understand. SBL is a busy time of the year. I may not be able to respond quickly with all that's on my plate as well.

[Stephen Carlson]

And to Dr. Carlson your objection to "summit" as a verb seems a bit disengenous since it could easily be substituted with the word "crest" which is recognized as a verb. It felt to me like a "gotcha" rather than a genuine engagement of his points. I'm sorry if I am overstepping my bounds in so stating this.

Accusing someone of being disingenuous is an accusation of dishonesty. I am strongly offended by it and I have reported the post.

[Scott Lawson]

Understood Dr Carlson. And I again apologize for possibly overstepping the bounds of propriety. But I was shocked by some of your comments to Mike. Again, I apologize if I misunderstood your intent.

My best regards.

Scott Lawson