In a number of places in the suttas, we find mention of various forms of canonical authority for religious tenets. The most famous example is the Kālāma Sutta (AN 3.63), but there are a number of other cases.
Two terms we find in such cases are itikirā (“so it seems”) and itihitiha (“so and so”). These have been usually translated by Ven Bodhi by “hearsay”, but that seems inadequate. Hearsay is mere rumor, and is not a basis for religious conviction.
There are a couple of things that can help us clarify this.
The first thing to note is that either one of these terms is used in combination with piṭakasampadā, anussava, and paramparā. This means that, as you’d expect, they have the same meaning.
The second thing is that in MN 95, there is the explicit mention of brahmanical scripture:
brāhmaṇānaṃ porāṇaṃ mantapadaṃ itihitihaparamparāya piṭakasampadāya
The ancient verse collections of the brahmins, the itihitiha, lineage, canonical authority
Note that sometimes, as here, itihitiha is compounded with paramparā, where it might mean “lineage of itihitiha” and at other times it (or itikirā) are separated from paramparā; but the difference is probably unimportant.
What is important is that these things are explicitly identified with the brahmanical texts, i.e. the Vedas and ancillary literature, as was known at the time. These are described in more detail thus:
ajjhāyako, mantadharo, tiṇṇaṃ vedānaṃ pāragū sanighaṇḍukeṭubhānaṃ sākkharappabhedānaṃ itihāsapañcamānaṃ, padako, veyyākaraṇo, lokāyatamahāpurisalakkhaṇesu anavayoti
He recites and remembers the hymns, having mastered the three Vedas, together with their vocabularies, ritual, phonology and etymology, and the itihāsa as fifth. He knows philology and grammar, and is well versed in cosmology and the marks of a great man.
Now, surely itihāsa here must have the same meaning. But here Ven Bodhi has “histories”; which is both inconsistent with his rendering the other terms, but is also impossible: history as we know—rational study of the past—it did not exist. In MN 76 he renders itihitihaparamparā as “legends handed down”, while at MN 95 he has “come down through oral tradition”. (This might be just a typo.)
“Legends” or “myths” is tempting; but the brahmanical texts as we have them are not exactly legends. The old texts contain or allude to various legends or myths, but they are not themselves collections of myths. The famous Hindu myths are only attested much later.
This is all a bit confusing, and I would suggest that we have one rendering for all these contexts, a rendering that makes sense in terms of the brahmanical scriptures that (so far as we know) existed at the time.
There do seem to be a couple of different senses at play, in that in the latter context itihāsa is just one aspect of the scriptures, whereas elsewhere itihitiha and itikirā are a generic terms for sacred texts. But this is not unusual. We have a similar usage in, say, the word “scripture” itself, which can mean either any sacred text or specifically the Bible. In Buddhist cultures, sutta is used in the same way.
It seems clear that the words here invoke some kind of ancient text (remembering, of course, that all such texts were oral). The terms themselves suggest the idea of a saying; the common element iti is used to indicate a quotation.
I suggest that we use “testament”. It is used of sacred scriptures, but it does not imply writing; rather it implies the “testimony”, i.e. a record of a serious statement of someone. And it’s not limited to the Bible, as we have “last will and testament”. It seems to work well in all these contexts.