Nornen there is absolutely nothing in Esperanto grammar that forbids estus + participle, and if you run a quick check on the Tekstaro you will receive many hits (over 1000), of which a high proportion will be estus -inta(j) (20 in La Faraono, 5 in Gerda Malaperis, 79 in Monato and Ondoj, to give a few examples).

The usage evidence is that estus -inta is not the preserve of English speakers.

Whether you call estus -inta a past conditional (or label it as equivalent to 'would have X-ed' in English is perhaps another thing. (By the way, Mi estus farinta could be either would have done or would have made in English.)

Unfortunately, the idea that you should use plain -us, regardless, is often trotted out in the forums, along with the dogma that -us is without temporal implication.

If I bring you coffee and you drink it and then say Mi ŝatus teon I get a whole different understanding compared to you saying Mi estus preferinta teon.

Yes, estus X-inta is regular, sometimes useful, and appears both in the Fundamento and in Zamenhof's works.

BUT: In all of these cases estus X-inta is paralleled to estu -inta, esti X-inte, estas X-inta etc., viz X-inta expresses something that has happened before estus.
And estus can be in past, present, or future, so can X-inta.

Misusing X-inta to express past only, as like speakers of (western) European languages, is a complete break of the lingustic system of Esperanto, a high cost that comes with little use.

If I bring you coffee and you drink it and then say Mi ŝatus teon I get a whole different understanding compared to you saying Mi estus preferinta teon.

Fakte mi preferus teon - Nun mi ŝatus teon. Much simpler and elegant.

I believe -us is often used to soften verbs like voli, deziri, esperi, demandi. You see a similar structure in Indo-European languages, things like "I would like" instead of "I want". "deziris" would be good here too, though.

Yes, Duck, but ....

Whilst 'who wished he was/had been a prince' conveys the idea of the soldier not being born a prince (highly relevant in the tale), 'who wanted/wished to be a prince' would suggest that it is not impossible for him to become a prince.

In Esperanto, kiu deziris esti princo, also allows the possibility. I don't at the moment see how you could get the required meaning of being stuck with commoner status, in a economical fashion, using deziris. (kiu tre deziris ke li estu naskiĝinta princo?)

Perhaps dezirus does the trick. It seems to go some way towards 'who would like to have been a prince, but not completely successfully.

The Tekstaro shows dezirus is often followed by ke -u, also occasionally by ke -us, even one hit with ke -is.

So - kiu tre dezirus, ke li estus princo; kiu tre dezirus, ke li naskiĝus princo.

Any better ideas ?

Kirilo81:Yes, estus X-inta is regular, sometimes useful, and appears both in the Fundamento and in Zamenhof's works.

If I bring you coffee and you drink it and then say Mi ŝatus teon I get a whole different understanding compared to you saying Mi estus preferinta teon.

Fakte mi preferus teon - Nun mi ŝatus teon. Much simpler and elegant.

Kirilo, the chap has drunk the coffee. Fakte mi preferus teon is baffling compared to the plain speaking of Fakte, mi estus preferinta teon.

You have only to look at the thousands of examples of the actual usage of -us to see that it has customary implications, never mind how much from a theoretical standpoint you argue that it is in principle timeless.

That's the fine thing with the Fundamento: Thousands of errors are still errors.

sudanglo:Unfortunately, the idea that you should use plain -us, regardless, is often trotted out in the forums

Really? It's often been said, correctly, that simple forms are preferable but I don't think it's been suggested it should be used without any regard for clarity.

sudanglo:along with the dogma that -us is without temporal implication.

You'll have to be a bit clearer on what you mean by "temporal implication", but what has been stated is that -US verbs do not show time per se, as do the indicatives. This is not dogma, it is relevant fact.

The basic problem is that rules or rules-of-thumb (of very fat thumbs) along the lines of "English past conditional is translated into Esperanto as estus x'inta, but you may use a plain x'us form if the meaning doesn't suffer" are plainly wrong.

Esperanto knows three absolute tenses: present, past and future. These are expressed in the indicative by -as, -is and -os respectively.
Esperanto knows three relative tenses: simultaneity, anteriority and posteriority. These are expressed in the participles by -ant/-at, -int/-it and -ont/-ot respectively.
Esperanto knows one conditional mode: -us. It is tenseless, it lacks both absolute and relative tense.

You may consult both PAG and MPEG that with composite forms, the copula expresses the absolute time and the participle expresses the relative time (relative to the absolute). Unfortunately "estus" lacks absolute time, and we have to draw this information from context, generally from neighbouring indicatives or adverbials of time.

Let's take this example:
(1) My father was a soldier and would have fought for his fatherland.

If I simply thought "Hey, 'would have' translates to 'estus x'inta'!" I would end up with:
(2) Mia patro estis soldato kaj estus batalinta por sia patrujo.

The conditional "estus" lacks time, so we have to infer it from context: the indicative "estis" yields "past". Now "estus" refers to the past and "batalinta" expresses anteriority. So my father would have fought for his fatherland before he became a soldier. Epic fail.

The correct translation would be:
(3) Mia patro estis soldato kaj batalus por sia patrujo.

Being a soldier and being willing to fight for his fatherland happen at the same time.

On the other hand side, the following example with "estus x'inta" does make sense:
(4) Mia patro ekestis pacisto kaj estus batalinta por sia patrujo

Again, "batalinta" expresses anteriority relative to "ekestis": My father would have fought for his fatherland, before he became a pacifist.

Bottom line:
Sometimes "would have" can be translated as "estus x'inta".
Sometimes "would have" must not be translated as "estus x'inta".
Putting "would have" and "estus x'inta" into direct relation is wrong and misleading especially for beginners.

----
Btw, frequent occurrences of "estus x'ínta" in the tekstaro just mean that "estus x'inta" is a valid, meaningful and well defined verbal form in Esperanto. It does not mean that "estus x'inta" is a past conditional. (The PAG states clearly the difference between past conditional and conditional perfect). Also the frequent occurrences of "iras" in the tekstaro doesn't imply that "iras" means "banana".

nornen:
(1) My father was a soldier and would have fought for his fatherland.

(2) Mia patro estis soldato kaj estus batalinta por sia patrujo.

(3) Mia patro estis soldato kaj batalus por sia patrujo.

(4) Mia patro ekestis pacisto kaj estus batalinta por sia patrujo

It is strange for me to see indicative and conditional, joined by "kaj", in one Esperanto sentence.

I would say:
(3) Mia patro estis soldato, tial li batalus por sia patrujo.

(4) Mia patro ekestis pacisto, kvankam li estus batalinta por sia patrujo

Can you find indicative and conditional, joined by "kaj", in Tekstaro or other Esperanto book?

sergejm:It is strange for me to see indicative and conditional, joined by "kaj", in one Esperanto sentence.

[...]

Can you find indicative and conditional, joined by "kaj", in Tekstaro or other Esperanto book?

As I have posted in this same thread, Mr Zamenhof wrote:

LLZ:En la nokto la hundo venis denove, prenis la reĝidinon sur sian dorson kaj kuris kun ŝi al la soldato, kiu ŝin tre forte amis kaj tre dezirus esti princo, por povi edziĝi kun ŝi.

The important point however is, that we two agree on the conditional forms:

sergejm:I would say:
(3) Mia patro estis soldato, tial li batalus por sia patrujo.

(4) Mia patro ekestis pacisto, kvankam li estus batalinta por sia patrujo

One should have a look at the other forms without temporal information in combination with participles:

estu X-i(n)ta: The volitive has a usual pragmatic reading of future, so X-i(n)ta can be anywhere in the past, present, or future, it just has to be finished before the time of estu (an example I used just today: La smartioj estu manĝitaj, kiam ni estos ĉe la infanvartejo - our son hadn't started to eat them at all).

esti -i(n)te: The infinitive often can be interpreted as timeless in the sence of being always true (gnomic use): esti manĝinte estas pli bone ol esti manĝonte - this is always the case.

Even more than in the combinations with -as/-is/-os, the time of X-i(n)ta with the other verbal forms depends of the contextual interpretation of -i/-u/-us.
And this leads to the paradoxon that, if estu X-i(n)ta were to have a past reading, this should come from -us, which in this framework however should be anything but not past.