Qanat | Hacker News Reader

Qanat

en.wikipedia.org

186 pointsbaghali69 comments

I first encountered this word reading Frank Herbert's Dune, so it was a bit odd when I encountered it in other literature. Later I realised one of the great things about Herbert's SF - he uses many real-world cultures as sources for his world building

Hah. I've just reread these, and I have been vaguely intending to look up 'qanat' to see if he made the word up.

Incidentally, I just came back from a holiday in Madeira --- amazing place if you like mountains; the average altitude of the island is 1300m --- and it's covered with a network of tiny canals called levadas. These collect trickles of water from the high summits, which are usually covered in cloud; it doesn't rain much. So they're kind of the opposite of a qanat. Instead of the water table being underground, it's way up high.

https://goo.gl/photos/a7eNGyJtiztiDFA86

Epicly, a lot of the levadas are in grooves chipped into the side of cliff faces. You can just about extrapolate the slope in that picture. Unfortunately in the really scenic bits I was too worried about not falling off to take photos. Safety railings are things which happen to other people in Madeira...

For me it was like having an insight into the subtext of the series. Quite a lot of Persian (PadehShah for king) , Arabic (Karama for blessings) , and Islamic (Mahdi for The Guided One), for example.

Yep. I've remembered it because it would be quite useful in Scrabble (it's in the OED), but so far I haven't had the opportunity to use it.

Indeed, now I always associate qanats with Leto Atreides II.

> he uses many real-world cultures as sources for his world building

Yup, like Kwisatz Haderach which is in Hebrew.

There was an interesting problem about qanats and excavation cost minimization in the 2015 ICPC World Finals problem set.

(Problem H) https://icpc.baylor.edu/worldfinals/problems/icpc2015.pdf

The section about using qanats for cooling and ice storage by the Persians around 400 BCE [1] is fascinating.

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

the last photo in the link was 10 min walk from my house. At that time (20yrs ago) they'd turned it into a ball pool for kids to play. Currently they've turned it into a Zoorkhaneh (https://en.wikipedia.org/wiki/Pahlevani_and_zoorkhaneh_ritua...). This is nostalgic!

Related technology: https://en.wikipedia.org/wiki/Windcatcher

Those look a lot like Dune's wind traps. Coincidence?

Considering that Dune is a metaphor for the Middle East and borrows heavily on Islamic history, maybe not.

Is there an etymological link between "qanāt" and "canal"?

The answer appears to be yes. Both Arabic qanāt and Greek kánna are related to the Akkadian qanû, meaning "reed". The Greek word became latin canna for cane, then canalis for pipe or channel. See this entry in the Etymological Dictionary of Arabic for more: https://www2.hf.uio.no/polyglotta/index.php?page=record&view...

It's the exact same thing.

For example, Suez Canal is called "Qanat al-Suez" in Arabic.

That doesn't answer the GP's question of whether there's an etymological link. That is, are the two words derived from the same root in some ancient language? It is a reasonable question since they are phonologically similar, but from different language families.

And just because two words sound similar doesn't mean they have similar roots. The word for “dog” in the Australian Aboriginal language Mbabaram happens to be “dog,” not because they have the same root but by coincidence.

https://books.google.co.uk/books?id=UtFqXQosVP0C&pg=PT283&lp...

Just to add to the fun, the etymology of the word "dog" itself is completely unknown.

There's a single documented mention of the Old English "docga" - and then nothing.

How this word took over pretty much well entirely from "hound" is a mystery.

If that's not enough, the Polish word for dog - "pies" - has exactly the same issue!

(and I believe, but can't say with any certainty, that other Slavic languages may have the same issue.)

Edit: Oxford English's page on "dog" is paywalled off but still available in via Google's cache, a fascinating read if you have a spare 15 minutes:

http://webcache.googleusercontent.com/search?q=cache:jNMSJSG...

https://en.wiktionary.org/wiki/Reconstruction:Proto-Slavic/p...

There's a great list of these at

https://web.archive.org/web/20070227031854/http://members.ao...

they're known as "false cognates" in linguistics: https://en.wikipedia.org/wiki/False_cognate

This one btw doesn't include the one that is strangest to me: that Chinese for mom or momma is 妈妈 "mama" and pop, pappa is 爸爸 "baba".

What is also interesting is that certain words share similar literal constructions in unrelated languages.

The best example I've always thought of is the verb "to understand". Picture standing under something; you get to know it better.

In my native language Twi [1], the to understand is literally constructed as "to to sit under" I know of several other languages with very similar constructions.

[1] http://www.omniglot.com/writing/twi.htm

These are nearly universal AFAIK. I've heard stipulated a link between this and the fact that these are the first sounds a child will learn.

https://en.wikipedia.org/wiki/Mama_and_papa

> these are the first sounds a child will learn.

Not really, no. 'Gaga' or 'kaka' are just as easy yet aren't universal. The mystery is not solved. (Could be just areal effects, for example.)

Easy for an adult, not so easy for a baby. Bilabial and coronal sounds don't take much tongue dexterity: you just need to be able to open/close your lips and leave your tongue in its natural resting position respectively to produce them. Velar sounds, on the other hand, require you to flex the middle of your tongue to touch your palate, which is as significantly more difficult sound to produce.

It's no coincidence that sounds like 'gaga' and 'kaka' tend to be associated with either more distant relatives or with defecation: the kid's going to be older when those things get introduced.

There's nothing even slightly strange about it. The terms are self-recapitulating because parents mistake children's babbling for first words. If anything, it'd be strange if they weren't the same. Take a read of this: http://languagehat.com/trask-on-mama-and-papa/

Addendum: in case anybody's wondering what I meant when I wrote "it'd be strange if they weren't the same", I was referring to the high probability of them being coincidentally the same or at least very similar.

Then why isn't aunt "gaga" in all languages?

Why would it be? What's your reasoning for expecting it to be? Did you read the paper linked to?

I always presumed these were universal because they are the easiest sounds for infants to make.

I'm not sure why that was downvoted. It's a reasonable assertion.

And correct, as it happens. Though long-running cultural associations with certain consonants (such as 'm', 'n' and 'b' and mother or grandmother, 'p' and 'd' with father or grandfather, &c.) to play a huge role in it too: the parents need to be culturally primed to mistake the babbling for actual half-formed words, and subsequently build that association in the child's mind in return.

Couldn't have said it better (obviously) myself. Thanks!

The 'ma' and 'pa' sounds are some of the easiest sounds to make by babies since they just require the lips and don't involve the tongue.

You might be interested in the article by Larry Trask linked to here: http://languagehat.com/trask-on-mama-and-papa/

I'm not sure why I was downvoted - Larry Trask's article has the exact same explanation i gave above - mainly that the 'm' and 'p' sounds are the easiest consonant vocalizations that babies can do.

canal deriv. : 1400-50; late Middle English: waterpipe, tubular passage < Latin canālis, perhaps equivalent to can (na) reed, pipe (see cane ) + -ālis -al1; def. 5 a mistranslation of Italian canali channels, term used by G. V. Schiaparelli

The linked article has an etymology section, says qanat is arabic for channel.

I am not an expert by any means, so we should wait the opinion of someone more competent than me, however.

I will say that there are not any link.

Most words in English come from two different roots, the German one and the Latin one.

It actually happens than in Latin "canal" is "canalis", so I can believe that the English word "canal" comes from Latin.

Usually Latin and Greek are the two roots and you don't explore deeper than that, but, given my limited knowledge, I cannot exclude that "qanat" comes too from Latin.

Edit: For completeness, the intermediated step between the Latin "canalis" and the English "canal" could be the Italian "canale"

Edit2: Actually from this page [1] it looks like the Latin "canalis" comes from the Greek "ka na" perhaps from the Assyrian "qanu". So it may be that there is a common root.

[1]: https://en.m.wikipedia.org/wiki/Talk:Qanat

> Most words in English come from two different roots, the German one and the Latin one.

Not quite?

What I'm going to say is not exact but as a first approximation it should be okay[1].

English is a Germanic language[2]. Most (all?) common words in Modern English share a common ancestor with their counterparts in Modern German/Dutch/Swedish/Danish/Swiss/Norwegian -- I've left out a bunch, sorry! So lexically and structurally they are very similar. That doesn't mean that most words are Germanic in origin, however it does mean that in any given text most word _occurrences_ are going to be. This accounts for a quarter (~25%) of the lexicon but the _bulk_ of words in any given text.

So words like "is", "can", "will", "must", "water", "the", "and", "word", "bread", "blood", and so on.

Of the rest English has borrowed from _many_ sources. Latin (but this is true of many languages) and French being equally predominant, a bit less than 30% each. Latin because of the Roman Empire and subsequently Latin being the language of European learning. French because of the Norman conquest of England. You ought not to think of this as being borrowed twice over from Latin, once directly, once indirectly. The influence of French on English is immense.

Of the remaining, (Ancient) Greek at a bit more than 5% features predominantly -- though it punches above its weight because so many crucial technical words are from this source, words we'd be hamstrung without.

Then words derived from proper names, then words derived from other European and Indo-European languages. Then the rest of the world.

As Persian/Farsi is an Indo-European language it would be unwise to say there is no link. Especially if the technology is old and the technology spread. And I think a sibling-ish comment points out that there is a direct link.

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

[2] http://www.theguardian.com/education/gallery/2015/jan/23/a-l...

You left out the Arabic influence, which is considerable for a huge amount of words in the sciences, math, chemistry, astrology, geology, botany and so on.

https://en.wikipedia.org/wiki/List_of_English_words_of_Arabi...

Well … I didn't single it out even though it is on the border of being statistically significant but then again I did not mention any Semitic language. I didn't technically leave it out; it's lumped together in the "rest of the the world" bucket. Going by the link you gave me there are ~150 words of Arabic origin in English excluding domain specific jargon, far less than 1% of the language. But I take your point -- probably coming after the languages I mentioned and taking into account the order I mentioned them in Arabic would come next.

As a sometimes Scrabble player, I was familiar with the word and I knew it was some kind of aqueduct. It's nice to know more. Amazing that ancient people not only knew to make them, but that they could pull it off. This must have been no small engineering feat.

Hard mode in Scrabble is giving definitions to all those weird words you memorise just for Scrabble....

the arabic word for them, FALAJ (and its plural AFLAJ) are also useful scrabble words (lowish probability, but very useful when you do have them playable)

Quite interesting how plural form is an simple anagram of singular form. Is it common in arabic or just a lucky coincidence?

I know absolutely nothing about Arabic, and stumbled across this page about Broken Plurals on wiki:

https://en.wikipedia.org/wiki/Broken_plural

Look a few rows into the patterns table, and you start seeing fun rearrangements for patterns such as CaCaC -> ʼaCCāC, the example being sabab -> ʼasbāb.

Pretty fascinating stuff!

It is one of the common plural forms. (In Arabic writing the words look even more similar, since it is only the three consonants, with diacritical marks above and below to represent the vowels.)

Such (ancient) technologies plus state-of-the-art computing and communication systems is prime material for science fiction novels about highly advanced civilizations.

I think we should strive towards that - use the wisdom of our ancestors to build zero-energy systems, powered by naturally occurring energy, controlled and enhanced by computers.

There are lots and lots of things that have already been invented in the past, we should just listen to our grand grandparents a bit closer.. After all, they built them for future generations, including us..

I'd add nature to that 'list'. There's so much we can - and have - learned from what sometimes appear to be dumb or primitive processes.

My first thought is: can I make something like this work in Dwarf Fortress?

Its funny to see this on HN! I was born and grown up in a warm area in Iran full of qanats and to me it seems such a natural thing. I caught my first fish from a qanat!

The Arabic countries led by the Muslims were the most advanced scientists/engineers in the world, until they let the religious crazies take over. https://en.wikipedia.org/wiki/Timeline_of_science_and_engine...

Being overrun first by primitive Mongol hordes that looted their wealth. This is like saying the Romans were advanced before the Christian crazies took over.

There is also this video about Qanats in Iran: https://www.facebook.com/32848633159/videos/1015351116587816...

Sorry for the fb link, could not find it in YouTube

The section of the wikipedia article on qanat's in Iran is huge (3700 words) in general, and especially compared to the sections about qanats in other countries. Between that and this video on qanats in Iran; what makes them so much more important/studied/etc.? Is it just that they were invented there?

> Is it just that they were invented there?

Well probably not just, but Qanats are persian tech and spread from there, even today the majority of qanats (both historical and extant) is in Iran, with still-active qanats more than 2000 years old. So it makes sense that this is where they have the largest cultural presence, and understanding.

Looking at the article's history it seems to have been a one-time contribution by an editor.

You might also want to learn about Karez that is used in Balochistan, Pakistan.

https://www.youtube.com/watch?v=lhpjbi93Jzg

My first thought (even though it's a bit of an overused meme) is that these look exactly like the holes in Japanese horror manga master Junji Ito's /The Enigma of Amigara Fault/ (don't look that up if you're claustrophobic or not cool with body horror).

Another article:

http://archive.aramcoworld.com/issue/201405/the.water.below....

Even better Scrabble-Word: https://en.wikipedia.org/wiki/Qat

I saw lots of these in Morocco, between the mountains and the Sahara.

Well, what I saw were the regularly spaced mounds of earth at the top of the access shafts.

I am trying to do pilot. Any good I can prefer?

We have them in Azerbaijan too.

ⵜⴰⵔⴳⴰ

I've watched a movie where they even have those in Palermo (Italy).

Reply on news.ycombinator.com