> The method you’re using is quite interesting.
> It’s inverse called a statistical language model
> (SLM) is used in speech recognition systems to
> determine the relative probability of one word
> following another.
I wasn't aware of this but it's quite apparent that statistics can be applied to vocabulary and language.
Just to be sure you're aware of it I should point out that I believe I've solved the intended meaning in the ancient writing. The Pyramid Texts is actually just the rituals read at ascension ceremonies and it's written in a different type of language than we use. It wasn't until long after I solved the meaning that I realized the proof is right before our eyes but it's invisible to us because of our expectations.
It appears from my work that the word order in Ancient Language was flexible. They didn't use subject > verb > predicate but rather subject > perspective > object while word order could vary. They also reasoned "backward" to how we reason and this affects word order as well. I'm not sure what the bottom line would be but suspect the results of such a study would be unremarkable.
Nobody knows how this language sounded.
> These models help the accuracy
> of a speech recognition engine by ruling out
> improbable results thereby reducing the search
> space and improving efficiency.
It should be possible to make many deductions using this process. Egyptologists believe these words carved into walls contain numerous errors and omissions including grammar, word usage, and spelling so would never apply such software. Of course I don't believe there are any errors of any sort in the PT other than transcription, publishing, and translation. There are likely two specific errors of science that reflect their limited understanding.
> The SLMs used are
> typically trained on thousands to tens of
> thousands of example sentences to deliver highly
> accurate predictions.
The PT is a significant corpus but much smaller than this and it repeats a great deal. Even utterances that don't look quite alike are restatements of one another. I know a great deal more about this subject than I've ever posted anywhere. The words repeat over and over because the language was like computer code and it didn't need may words. Almost all of the words were nouns. Even words like "nehebkau" which we believe is an Imaginary Consciousness was actually just a noun that meant "hydraulic cycle". Of course the concept of the hydraulic cycle representing the subject of a sentence takes on aspects of a verb.
> The method therefore has merit based on the
> original Heiroglyphs, a consistently applied
> transliteration or starting from a known or
> established set of translated Heiroglyphs from
> within the PT.
The translations are too far wrong to use them for such an analysis. Colloquial forms of verbs (walk, run, fly) are the closest to be translated exactly right. But even these depend on perspective to create meaning. It's hard for modern people to see how a star can stride or a cow can fly. But such things were important to establish context, perspective, and consistency in a sentence.
> It feels like there is a thesis
> work that could be done here applying AI models to
> Heiroglyphs known from multiple texts from that
> era. To do this would require the Heiroglyphs
> along with certain metadata such as estimated time
> of writing.
For practical purposes there is only the single text. Other work is from later eras and written in modern language. Even this one text that survives in Ancient Language was heavily redacted after it was no longer used for its original purpose; it was already on the road to the change that would eventually produce the book of the dead. A discontinuity occurred and then it became the book of the dead.
Curiously I believe Ancient Language could probably be directly adapted for use as a computer program because it is so similar to computer language already. I'd be inclined to design a computer from the ground up with reconfigured hardware but certainly a programmable computer would suffice. It would be hard to program.
> I’m not saying here that the result you’ve
> claimed is supported just the method is
> interesting and if applied methodically from a
> well laid out base has the potential to
> demonstrate something new and important.
There's a lot in this work. Scholars could be tweezing out more information for centuries unless we really find the Book of Thot which would take precedence over a heavily redacted silly little book of rituals.
Your method will be far more important to cracking the Book of Thot. I'm not sure I would understand all of it even if properly "translated" to reflect author intent. The science will be complex but the metaphysics will be far more complex.
> Outside of this, one of the major claims you make
> is that superstition kills ancient man.
Superstition kills everything and will even kill things around it.
It would have killed all ancient people as well if they were superstitious. We are highly superstitious and many die because of it but the species (homo omnisciencis) survives because we have agriculture and technology. There are extinction threats on the horizon in the next century and our superstitions might be our undoing.
Superstition is belief and all modern people have beliefs imparted by modern language.
I've long said a statistical study of the PT would be very eye opening. For all practical purposes I solved the meanings of the words using statistics. I don't normally perceive it this way as I'm working because most of the solutions are largely intuitive and achieved after learning all the possible configurations of the words.