phillies wrote:
Thought I'd have some fun with this...
(snip)
And it was good fun. But: Pronunciation is irrelevant for translating a written text in an unknown language.
More below.
AncientMariner wrote:
Phillies: is it safe to assume your Spanish section came out out of Google Translate or something similar? I ask because I see some typical automated translation problems in the section.
I confirm. My active command of Spanish is dismal. But I couldn't resist using that specific (English) phrase, even though I figured that it might not work out in Spanish. Anyways, it was intended as a demonstration, only.
AncientMariner wrote:
Both of these posts illustrate why I think the SSK would have trouble translating Seijin Khody's Spanish sections without any other context.
(snip)
If Khody was using Spanish colloquialisms and phrases, which have inherent meaning only to native speakers, the chances of successfully translating those sections of the journal would be very low. It's not enough to be able to map words from one language to the other. There are other considerations when assigning meaning and context.
I agree that it's difficult, but they have several centuries, likely a rather large body of text from one writer, and context from the English parts. So, while I acknowledge that can't come up with a full translation (e.g. for figures of speech), they should be able to figure out quite a lot about the content.
If the SSK would have a go at translating the Spanish sections, the following points should be relevant.
- Spanish is quite strict in terms of syntax. Usually, it's subject verb object (similar to English). Grammatical analysis allows for identifying verb forms.
- Both Spanish and English use conjugation with auxiliaries. You can then identify those composite forms. Assigning tempus requires context. But often simply knowing that it's a verb helps a lot...
- You can rather easily spot most adverbs (realmente, actualmente, ...) by equating “-mente” to “-ly”
- For adjectives, you have a sufficient number that end on -(c)al and are partly written quite similarly to their English counterparts like e.g. similar (only with pronunciation you could identify the Language I'm writing that word in...). Then you will recognize that adjectives usually go after their nouns, which helps a lot.
- Regarding articles, "la", "las", "el", and "los" you can equate to "the" by frequency sampling and word order. After that, you can identify a lot of nouns.
I could continue with more examples. Grammatical analysis will allow for identifying a lot of the actual Spanish grammar even if you don't know the meaning of all the words. And I posited that grammar is known on Safehold and should be accessible to the Sisterhood as well.
Now, for the meaning of words, you will build an ordered list of words. Spanish and English are based on using stems, and the respective rules are comparable. Thus, you can analyse the word list for stems, being aware that stems have (partially widely) different meanings (e.g. "set").
And then you will start to notice that stems in English and Spanish are partly identical (apart from a somewhat different spelling). For instance "realmente", you've just reduced to "real" and “-mente” and tentatively trying "really" will bring you a long way. Then, if you run into "la realidad" going for "the reality" is quite simple as you know it's a noun. Then, equating "realizar" to "to realize" is obvious.
That way you get further insights into word generation in Spanish.
Equating things like "seccion" and "section" after identifying the plural "-es" opens up more easy matches.
Then, there are prefixes. Some common prefixes ("re-") are identical or ("des-" - "dis-") similar. That'll give you access to more stems (in Spanish and in English) for comparisons.
Unless Safeholdian English was rather systematically (sistematicamente) purged from Latin stems and tailored to Germanic/Nordic roots (skin, kill, heaven, live, etc.), you have lots of roots in standard English to reconstruct (reconstruir) those Spanish words. Actually, the largest problems will be some arabic leanwords in Spanish, the fusion of Latin "v" and "b" to "b" in some cases, and the shift of Latin "f" to "h" (or deletion) in the basic vocabulary. And of course those irregular verbs like "ir", "ser", etc.
My experience is that word to word mapping gets you quite a long way on the basic meaning of a text (if you know possible meanings). Combined with basic grammar you get quite acceptable results. And as a bonus, you have a lot of context to draw from.
As stated above, there will be uncertainties (e.g. for figures of speech), but that should not prevent the Basic understanding of large parts of those texts.
If the SSK would try to read those texts, though, it would get funny - but not as funny as my attempts at spoken Spanish
.