Should have

Question

Hi,A Chinese proverb says, "When you are dying of thirst, it's too late to dig a well."  Carl Sauer argued long ago that starving people have no time, energy, or resources; they cannot invent agriculture or develop new crops.  He proposed that agriculture must have started among reasonably affluent, settled people.  This may or may not be so, but at least we can be sure it did not start among the truly desperate.  Hunters and gatherers are not as impoverished as many writers still imply.  When they do face want, they usually move, a strategy that makes farming even less attractive than it is in good times.  Thus farming probably started among people who had enough food; they presumably wanted to produce their favorite foods closer to home.It says in my english test book(written by a non-native) that a close paraphrase of the underlined sentence is as follows:Agriculture should have begun with considerably wealthy people.Q1) Do you think an epistemic reading of "should have" is possible in the above sentence?  The author of this test book seems to be claiming that the event expressed in the sentence may or may not have happened and that "should have" in question is a weaker version of "must have".  Would you agree with this?Q2) Sometimes "should have" is used with a past time situation where an event was expected to happen, but in fact it didn't happen. ex) He should have been home by now.(You expected that he had arrived home, but just found out he's not)Is this another epistemic use of "should have"?  Or it's more of a deontic use?

CalifJim · Accepted Answer

jooneyQ1) Do you think an epistemic reading of "should have" is possible in the above sentence?

Not in the variety of English that I speak.

jooneyand that "should have" in question is a weaker version of "must have". Would you agree with this?

No. It doesn't work that way in the English that I speak. (AmE)

jooneyQ2) Sometimes "should have" is used with a past time situation where an event was expected to happen, but in fact it didn't happen.
ex) He should have been home by now.(You expected that he had arrived home, but just found out he's not)
Is this another epistemic use of "should have"? Or it's more of a deontic use?

Your parenthetical remark focuses on expectation, so it's epistemic. Absent your gloss on the situation, however, I would have said that it's impossible to tell from the isolated sentence. It seems to me that if he had been instructed to be home by now and had failed to obey instructions, the statement would more likely be heard as deontic.

CJ

Dave_anon · Answer

Hi

I would use the word deontic to describe how people ought to behave, according to their society, law or religion. In such a case, "must" and "should" work in the same way..

- If he committed the crime, he should admit it to the police

- If he committed the crime, he must admit it to the police

These two sentences, I would say, mean the same thing

However, you raise a good point when the words are used epistemically. If the conclusion of an argument is postive then the word "must" is usually used. When the conclusion of an argument is negative then "should" is usually used

- If 'a' is equal to 'b;' and 'b' is equal to 'c' then 'a' must be equal to 'c' (and it is!)

- If frogs are mammals then they should be warm-blooded (but they aren't - so they aren't mammals)

So, for (Q1), I too would query the words in the text book...

- In order to develop an agricultural system, you must already be well-fed, and that is what we find

- If only well-fed people can develop an agricultural system, then we should not find desparate people owning farms - but we do

If your translator is agreeing with the argument - which I think they are - they should say..

- Agriculture must have begun with considerably wealthy people

If they were arguing against the case, they might say something like..

- So then, agriculture should have begun with considerably wealthy people.- but this is not so

So, yes, I agree with you also about (Q2)

However, the sentence you use there can be deontic or not..

- He should have been home by now (because I told him he must be home by 9 o'clock)

- He should have been home by now (because the train usually gets here by 9 o'clock)

But I don't think (Q1) depends on deontism - I think it depends on whether we are expecting a yes or a no answer

Hope this helps - do get back if I can clarify it

Dave

Jooney · Answer

Thank you very much, CJ and Dave for your answers.

Dave, could you explain this part in more detail?

"If they were arguing against the case, they might say something like..

So then, agriculture should have begun with considerably wealthy people.- but this is not so"

You seem to be saying an epistemic interpretation of "agriculture should have begun with..." is possible in a certain situation. Could you explain more?

Jooney · Answer

Agriculture should have begun with considerably wealthy people.

You say this when you go against the argument that agriculture started among relatively poor people, right?

By looking at all the evidence, I disagree with your statement and must conclude that it was the affluent people who started agriculture, not the poor. Did I get that right?

Jooney · Answer

Sorry for making a series of questions.

He should have been home by now.

When you can say this as a way of expressing conjecture of a past time situation, why can't you do the same thing with the agriculture example?

Agriculture should have begun with considerably wealthy people.

I'm not 100% sure, but my guess is that it is probably true that agriculture started with rich people.

I may be missing something here. Please help me understand this.

CalifJim · Answer

jooneyI may be missing something here. Please help me understand this.

You may be missing the fact that 'should have's are counterfactual.

Earlier, you said that the original statement

Thus farming probably started among people who had enough food.

could be paraphrased (according to your book) as

Agriculture should have begun with considerably wealthy people.

I am assuming, rightly or wrongly, that they intend to paraphrase 'people who had enough food' as 'considerably wealthy people' (as oblique as that may be as a comparison). Moreover, they intend to paraphase 'farming' as 'agriculture' and the verb 'start' as the verb 'begin'. Thus, they are establishing the following equivalents:

[Farming / Agriculture] probably [started / began] [among people who had enough food / with considerably wealthy people].

~

[Farming / Agriculture] should have [started / begun] [among people who had enough food / with considerably wealthy people].

The crux of the matter is whether probably (started) can be accurately paraphrased as should have (started). What do you think? I think it's clear that if we say that something "should have started", it did not start at all. This is the counterfactual nature of "should have".

Let me interrupt here with a reference to the other sentence you mentioned. Note that He should have been home by now implies that he is not home now. This is what I mean by 'counterfactual'. In general, anything that "should have happened" is precisely something that "has not happened".

Returning to our main problem. When we say that something probably started, do we mean that it did not start? In other words, does the use of 'probably' make the statement counterfactual? No. Not at all. In fact It probably started shows some confidence that it really did start.

So how can a counterfactual expression be a paraphrase for an expression that is not counterfactual? I'd say that that is impossible. Nothing that implies "It did not start" can possibly be a paraphrase of "It probably started".

CJ

Jooney · Answer

Thank you CJ for the explanations. They are really helpful.

Here is one thing that still bothers me.

According to Practical English Usage by Swan, "'should have' can be used to talk about past events which did not happen, or which may or may not have happened." He gives following examples.

ex1) I should have phoned Ed this morning, but I forgot.

ex2) Ten o'clock: She should have arrived in the office by now.

I can easily see ex1) is a counterfactual situation, but what about ex2)?

Swan seems to portray ex2) as an event that may or may not have happened. Doesn't he?

CalifJim · Answer

jooneyex1) I should have phoned Ed this morning, but I forgot.
ex2) Ten o'clock: She should have arrived in the office by now.
I can easily see ex1) is a counterfactual situation, but what about ex2)?
Swan seems to portray ex2) as an event that may or may not have happened. Doesn't he?

Yes, that's true. I hadn't thought of that case. In the second case the speaker is guessing about a situation he has no direct knowledge of. He's not there so he doesn't know if she has arrived in the office or not.

Nevertheless, I still can't see how the 'may have happened or may not have happened' scenario can apply to the agricultural story. 'may have started or may not have started' doesn't seem to me to be a paraphrase of 'probably started' any more than 'didn't start' is a paraphrase.

If I think of an explanation that is more satisfactory, I'll post it.

CJ

Jooney · Answer

Thank you CJ.

I read the passage over and just noticed something. This author of the passage seems to be trying to draw his conclusion of a past event by looking at current evidence. He doesn't directly say he can't find any poor people running agricultural farm or land today, but implies that is the case given that starving people just don't have energy or time or resources to do such thing.

Then what about the 'home' example?

She should have been home by now.

Does the speaker express his opinion based on his knowledge of current situation? I just saw her car parked in front of her house. She must have come home from work. Is this what the sentence means? Is this the way "should have" works when you guess about what happened in the past? What do you think CJ?

p.s. I'm asking for your opinion. I'm not trying to be assertive or anything.

CalifJim · Answer

jooneyShe should have been home by now.
Does the speaker express his opinion based on his knowledge of current situation?

My first instinct is to say 'yes', but it's also knowledge of the schedule established in the past. I'd say that, without further context, speaker knowledge consists of "She's not here at home now" and "She was scheduled to be home now". And the speaker may be worried because the previous arrangement that had been made assigning the time of the return home.

jooneyI just saw her car parked in front of her house. She must have come home from work. Is this what the sentence means?

No. These two sentences make sense as they stand (car parked ..., must have ...), but they don't add up to She should have been home by now. It's the 'have been' that creates the problem. From the evidence of the presence of her car, you can conclude either weakly, She should be home now (I expect/suspect) or strongly, She must be home now (I am certain). But you can't make any conclusions from the evidence that she should have been home at any particular time.

jooneyIs this the way "should have" works when you guess about what happened in the past?

No. (I'm having a little trouble following this, but I think you're referring back to the case brought up by Swan, She should have arrived in the office by now.) In any case, when you guess about what happened in the past, which is a relatively less common use of 'should have', you generally do it in the absence of current evidence. I think that's why seeing the car parked there doesn't fit the scenario you're trying to create. It's evidence, so you're not guessing about the past. You're guessing about the present.

Anyway, in She should have arrived in the office by now, you're saying that you know something you learned in the past, namely, that there was a certain time she was scheduled to arrive. However, you are not there at the office, so you have no current evidence that she's there; you're just guessing on the basis of what you know about the schedule discussed earlier. It seems to me that in all these cases you have to know something about a schedule established previously and not to know anything directly about the current situation.

Suppose someone is going on a trip which will include a stop in London (Tuesday, Wednesday). You never receive any communication from this person that would indicate where she is. On Tuesday, you can say any of these, and they all mean about the same thing:

She should be in London (by) now.

She'll be in London (by) now.

She must be in London (by) now.

She is probably in London (by) now.

I think she's in London (by) now.

She should have reached London (by) now.

She will have reached London (by) now.

She must have reached London (by) now.

She has probably reached London (by) now.

I think she has reached London (by) now.

You can't say, with the same meaning:

She should have been in London (by) now.

because it implies She has not succeeded in reaching London yet.

I don't know the exact reason why, but I suspect it's that stative verbs and dynamic verbs act differently in this grammatical pattern. I suspect you need a dynamic verb to convey the 'maybe, maybe not' or 'guessing the past' usage of 'should have'. We are getting into areas now that require explorations of academic papers on these topics. And that means I'm at the end of my rope and will have to stop for now.

CJ

Should have

Top answer

Get the Weekly English Kit 📬

Related Questions