A lost language encoded in Inca cords | Hacker News Reader

A lost language encoded in Inca cords

newscientist.com

66 pointsmikemoka32 comments

Is pretty lucky that us visiting Europeans didn't know there could be writing in them, otherwise we would have burnt the lot long ago, like we did with the written records of the Maya.

Walking around on the surviving remains of Maya and Aztec cities is quite impressive.

As European it saddens me how we almost wiped everything out back then.

From what I have read, there are maybe 14 surviving books out of the hundreds of thousands that we took. Possibly less since that museum fire. We destroyed 2000 years of an entire continents written history. And many people simply do not give a shit.

Was talking to someone who identifies strongly as Christian about this recently. She said that she didn't care what had happened to the Maya, because these people practiced human sacrifice. I had to point out to her that you do not get to complain about human sacrifice after committing genocide.

Human sacrifice was practiced by the Aztecs too of course (don't know about the Incas, but it was clearly an accepted and unremarkable part of life for a large region of the new world); the revulsion that it provoked in the Conquistadors was surely a significant factor in enabling them to convince themselves that destroying these cultures was justified. But I suspect that the conquest would have played out very similarly even if the Europeans hadn't been confronted with pyramids caked in human blood.

Well just look at the Inca. They didn’t do human sacrifice.

The Inca did indeed practice human sacrifice.[0]

[0] https://en.wikipedia.org/wiki/Mummy_Juanita

True, but the scale on which the Inca practiced sacrifice is not exactly comparable to the Aztecs, e.g [1].

1: https://phys.org/news/2018-06-skull-tower-rack-evidence-azte...

In The True History of the Conquest of New Spain [1] (a first-hand account of Hernán Cortés' expedition in Mexico, that brought down Monetzuma's Aztec empire) there is a passage where the author, discusses the matter of human sacrifice and cannibalism with some local priests (I think these are not Aztec, but Tlascalan priests, from one of the cities allied to Cortés).

I can't find the passage in my hard copy of the book, but, from memory, the priests basically state that offering human sacrifices to the gods is the pious and proper thing to do. Castillo ends up persuaded of this, although he continues to abhor the acts they perform, of course. These priests also practiced self-mutilation, so they had noses and fingers, etc, missing. I think the point was that they were offering parts of themselves, too, as sacrifice to their gods.

There's no doubt that human sacrifice is a barbaric custom, that was historically practiced by few societies in the same scale as it was by the Aztecs. However, it is important to understand that, to them, it was not an act of evil, or even of righteous revenge towards their enemies. It was an act of worship to their gods.

My point is that we cannot apply our own concept of morality, which is probably influenced by the dominant religion of the western world, Christianity, to the manner by which people lived their life in a society that might have, for all intends and purposes, been on another planet for most of its history. If we are to judge those people, it must be on their own terms. And on their own terms, human sacrifice and cannibalism was what their faith (and therefore, their morality) required of them.

_____________________

[1] https://en.wikipedia.org/wiki/Historia_verdadera_de_la_conqu...

[2] Bernal Díaz del Castillo, https://en.wikipedia.org/wiki/Bernal_D%C3%ADaz_del_Castillo

Btw, Bernal del Castillo's account is a beautiful tale. Real life doesn't come closer to epic fantasy than this. There's wild, mysterious lands, powerful empires, barbarian heroes, battles between armies of thousands of men, even magic, in a sense, in the awesome power of the gunpowder weapons and the never-before seen horses they rode, which must all have looked sueprnatural to the Mexicans.

It's on project Gutenberg, online:

http://www.gutenberg.org/ebooks/32474

>There's no doubt that human sacrifice is a barbaric custom

Our own culture believes in sacrificing humans to the things that we worship. We just dress it up differently.

>"The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants. It is its natural manure"

- Thomas Jefferson

Think of all the lives that went into building the beginnings of the industrial age: miners, construction workers, pedestrians electrocuted or run down by automobiles. Even today this continues in slightly different ways - as one example, thirty-seven thousand of our fellow citizens here in the US died on our roads last year. And we just think to ourselves that's just how it works.

And that's before getting into our cultural celebration of heroic sacrifice, or other ways we trade in human lives, both our own and others.

> We destroyed 2000 years of an entire continents written history.

The Australian aborigines have been around for 60,000 years. Their oral history has survived for 40,000 years.

We can't even maintain an oral history in the so-called West for 80 years.

Sometimes its a wonder we are so sophisticated at all. I guess going to Mars makes a difference. But often I wonder if in fact the oldest cultures on the planet aren't, possibly, the wiser ..

>us visiting Europeans

That's a rather gentle sounding euphemism for what really happened. Your statement may have been tongue-in-cheek though.

It is usually best to assume that I am indulging myself in facetious sardony. It is a moral failing on my part, but it keeps me entertained.

Unfortunately many thousands of quipus were actually destroyed after the Spanish conquest, because it was known they contained information.

The organization of the Incan Empire was really incredible. There really was nothing quite like it until modern times. The state storehouses alluded to in the article can still be seen all over on mountainsides in Peru (they were always built off the valley floor where humidity is lower). These stored so much grain that the Incan empire really didn't have famines. After conquest, the Spanish enslaved the Quechua and generally had them mine instead of grow food. The grain in the silos fed them for several years.

I have to think that quipus can encode non-numeric information to make something like that possible.

>If we are to judge those people, it must be on their own terms. And on their own terms, human sacrifice and cannibalism was what their faith (and therefore, their morality) required of them.

So, you're ok with the Spanish wiping them out? According to the mores of the time, Spain's actions were perfectly acceptable.

Thank you for identifying this most insidious of tropes: exculpatory relativism for everyone else, but harsh judgment for Western history. I can’t help but conclude that a profound Western self-loathing underlies it.

If that bit about "self-loathing" refers to me, you are way off the mark. I have no reason to feel guilty about collonialism. I am Greek and my people never egnaged in it. Instead, we were ourselves enslaved by the Ottomans during the time that other Europeans raided far away lands, killed their people and nicked their stuff.

Put that in your assumption pipe and smoke it.

To be fair, I omitted another thesis which on reflection actually seems more likely: a vilification of the victors qua victors. It seems that all else being equal, we will condemn one murderous conqeusting people while exculpating another, simply because the former triumphed over the latter.

Excuse me, but I am very confused. Which murderous conquesting people am I exculpating?

>> According to the mores of the time, Spain's actions were perfectly acceptable.

That is a deeply ahistorical statement. In their own time, the conquistadores were strongly criticised for their atrocities, particularly on the side of the church. For a most prominent example see the extensive Wikipedia article on Bartolomé de las Casas:

https://en.wikipedia.org/wiki/Bartolom%C3%A9_de_las_Casas

Also, try to think about it this way: if everyone was OK with collonialist era practices, why are we not OK with them today? What changed, and when did it all start to change? Do things ever change spontaneously, with out explicit criticism of current morality, and purposeful attempts to reform it?

(No, they don't).

There are always dissenters. I guarantee that plenty of people in Mesoamerican society did not agree with the conquesting and violent ways of the Aztec Empire.

I don't understand the relevance of your comment to my remark about De Las Casas, who btw was a reformer, not a dissenter [1].

I think you may not be replying to the content of my comment, but instead identifying it with some other, third party's opinion, that you wish to disagree with.

______________

[1] His criticism of the savagery of the treatment of the Indians in the colonies led to legal reform that improved their condition significantly:

https://en.wikipedia.org/wiki/Bartolom%C3%A9_de_las_Casas#Th...

Hardly just a bit of dissent.

This comment breaks the site guidelines. For example: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."

That guideline is there because not following it leads to flamewar. Please review https://news.ycombinator.com/newsguidelines.html and follow it and also the others.

We detached this subthread from https://news.ycombinator.com/item?id=18542311 and marked it off-topic.

This data is begging to be analyzed by machine learning algorithms.

I just happen to be reading Clive Cussler's Inca Gold (1994) where they use computers to decode a quipu cord from scratch in about an hour and a half. I'd love to see the spreadsheet used for the initial insight.

This is truly brilliant work. There must be other clever potential rosetta stone potential as the smart people in the article figured out.

I actually love these kind of trashy archeology-themed books, where the brilliant linguist decyphers the ancient writing system over his/her lunchbreak, but it's rather spoiled by the inconvenient fact that the actual process - mountains of data, years of slog - is so unsexy by comparison. Plus, there is never a Rosetta Stone (that would make it too easy!), whereas almost every real-world decipherment has needed something like one. The only exception I can think of is Linear B; I've always thought it'd be a fascinating exercise to repeat Ventris' painstaking cataloguing and cross-referencing on a computer to see if his breakthrough could be repeated more quickly, or automatically.

I don't see how machine learning can be of any help here. Were would the training examples come from?

To train a machine learning algorithm, presumably a machine translation system, you'd need examples of already translated khipu. It's clear from the artcile that we don't have anything like that. We know how to interpret numbers, we know where on a khipu they used to represent someone's clan, but we still can't read the things. What "data" will the machine learning algorithms analyse?

A lot of the work the article talks about seems to involve manually recording features like the materials and colors used, the position and type of knots and so on, followed by searching for similar patterns in other khipus. Machine learning could conceivably speed up that encoding work.

Given a large enough collection of transcriptions, one could use "sophisticated" techniques like counting ngrams to try and identify common Quechua words. Depending on the content of a single khipu, i.e. whether it's more like a sentence or a longer text, and how similar modern Quechua languages are, it might even be possible to leverage that into a somewhat usable (i.e. sometimes correct) machine translation system.

So you'll model the features. Then what?

Counting n-grams will tell you nothing useful, unless the language is Quecuha, and a dialect that is well known, at that. But if it is such a dialect, why are we not any closer to understanding it? You don't need machine learning for that, or even a computer; it helps, but the work can be done by hand (see The Golden Bug: https://en.wikipedia.org/wiki/The_Gold-Bug).

For khipu that are clearly not narrative, i.e. include mostly numerical, plus some other, possibly literary information, it would be even harder to pull off. You'd have to understand words out of context, including names etc.

At best you might end up modelling the structure of an unintelligible language- like I think was done for the Voynich manuscript. You'd still not be any closer to undestanding it.

Reply on news.ycombinator.com