sudanglo:I didn't find it documented in Murphy's English Grammar in Use - nor in Swan's Practical English Usage. Perhaps you could give the exact reference.

I've got the blue Murphy's book for intermediate students right here. "English Grammar in Use", Unit 48 (p96), part A.

sudanglo:However, what I saw in Swan did suggest a sentence which I might find convincing (as, perhaps, is your Italy example).

The man on the TV said that birds have been around since the time of the dinosaurs

That's clearly a good example of the present perfect being valid in such a sentence.

sudanglo:By the way, since '-is' in Esperanto can function either as English's Simple Past or as its Present Perfect, it seems quite legitimate to add the 'tiam' to force the meaning of the Past Simple - which implies a THEN in English.

I was talking about English grammar, not about Esperanto grammar.

Thanks for that Chainy. I appear to have the same edition - Unit 48 is on page 96 in my copy.

But what it says there about not using the past shift in reported speech is illustrated with examples of not changing the Present Simple to the Past simple. A point not in contention. (Ann said that she wants to go to New York next year. is a good example)

[It may be relevant here that the Present Simple often has a 'sentempa' quality, whereas the Present Perfect is more firmly linked to the time line.]

Anyway, I think you are right in distinguishing two threads in the discussion. The issue of pedagogy - providing a cogent guide to the foreign leaner - and the issue of whether a certain usage can be found.

Swan provides a striking number of cases, he has actually found, of the Present Perfect being used with an expression designating a time in the past - in contravention of the Present Perfect's meaning. But he cautions, against this for the foreign learner.

And he is right to see this usage as not legitimate - having a careless or arbitrary quality.

The question for me was whether the use of the Present Perfect in reported speech was legitimate - could be coherently accounted for, and what the function was. Not the question of whether actual cases could be found in the speech or writings of the native speaker.

I concede, now, that it has a validity, and I think I understand the principle of when it is valid, but I contest that you can find it accounted for in any textbook for foreign learners.

sudanglo:Thanks for that Chainy. I appear to have the same edition - Unit 48 is on page 96 in my copy.

But what it says there about not using the past shift in reported speech is illustrated with examples of not changing the Present Simple to the Past simple. A point not in contention. (Ann said that she wants to go to New York next year. is a good example)

The significant line in the Murphy's book is:

It is not always necessary to change the verb in reported speech. If you report something and the situation hasn't changed, you do not need to change the verb to the past.

The Murphy's book that we are talking about is for intermediate learners of English. That is why it gives only very limited examples of the present simple remaining unchanged in reported speech. But this same rule also applies for the present perfect. I don't have a copy with me right now, but I'm very sure that this is clearly explained in the book "Advanced Grammar in Use".

sudanglo:The question for me was whether the use of the Present Perfect in reported speech was legitimate - could be coherently accounted for, and what the function was. Not the question of whether actual cases could be found in the speech or writings of the native speaker.

I think the various examples that we have looked at make this perfectly clear.

I've found a pdf copy of "Advanced Grammar in Use" by Martin Hewings. The relevant section is in Unit 45:

I still prefer 'the crushed victim agreed that his opponent had won' - not 'has won' even though your having won is unchanged.

I apologize in advance for adding further comment to this thread. It's sort of been done to death.

However (at least for me) there is one interesting loose end.

I don't remember now who brought it up - it might have been you, Chainy - but the idea was floated that the interval between now and the occasion of the actual direct speech might be relevant.

I think it is quite clear that in the case of the Present Simple in the subordinate clause (when it is not past-shifted) this interval isn't relevant.

I can equally say 'She told me a few minutes ago that she wants to go to New York' or 'She told me a few weeks ago ...' or 'She told me last year ...'.

Provided I wish to communicate her current desire, the absence of past-shift is allowable.

BUT, it is not clear to me whether the interval between reporting and actual speech functions in quite the same way with an un-pastshifted Present Perfect in the subordinate clause.

THOUGH if the justification is COMPLETELY accounted for by the idea of the currency of the matter in the subordinate clause, then it would seem to be reasonable to conclude that it doesn't matter when the actual words were uttered.

Yet there does seem to be a difference between the acceptability of the reply to Chainy's snoozing gentleman 'He said he's been to Italy 3 times' and (for example) 'He told me last year he's finished the book'

It might be worth a separate thread to explore whether a comparable complication exists for Esperanto.

sudanglo:It might be worth a separate thread to explore whether a comparable complication exists for Esperanto.

Maybe - it got a bit stressful for me trying to participate in this discussion knowing that I wasn't really being on topic.

But I think things have been fairly well wrapped up here, between you and Chainy.

I don't think there's a comparable complication for Esperanto though, unless we're using complex tenses.