6 ms·

Given a failed, decades-long effort to decipher Linear A, how likely is it that we will ever be capable of deciphering a purely alien transmission?

sipior 4y ago
I think the only saving grace there is that the sentients sending such transmissions would presumably want to be understood, and would ensure a Rosetta Stone of sorts was packed along with the message; the common element of reference probably being the mechanism of the transmission medium itself. Contrast that with the Minoans here, who were just writing for other Minoans. But given how much culture and biology is inevitably encoded into even our human languages, one does sometimes wonder if we're even asking the right questions. I hope I live long enough to see one of the various SETI initiatives pan out into such an amazing discovery. As a child, I thought it inevitable; pushing fifty, I'm rather less sanguine.
showerst 4y ago
We _might_ share a set of universal constants with any other intelligence that they also care about. PI, the speed of light, binary, etc. In a weird sense specific to decoding transmissions we might share more of interest with them than Minoans, who were mostly likely recording the names of gods/heroes and accounting records for commodities that may no longer even exist.
- ncmncm 4y ago
  Tau, anyway. It seems unlikely anybody else would make the pi mistake.
Retric 4y ago
One big hope is it would include video to give context. Volume of text and consistency should also make decoding easier. Our Linear A samples are mostly fragments without context and they contain various inconsistencies.
- jhauris 4y ago
  I don't think sending video is a likely scenario if the sender wants to be understood. There are too many ways video could be encoded, even assuming the aliens have the same perceptual capabilities we have. Flipping it around, if we were going to do send a message to the universe would we send an h.265 bgr video with aac audio in streaming mp4 format? The only thing you can know for sure you have in common with someone receiving your radio signal is that they can receive radio signals, so using something about radio transmission would be a good bet for building a lexicon. Then there's the universal constants and mathematics which we could assume are known in common.
  - Retric 4y ago
    Uncompressed video formats are vastly easier to decode than h.265 bgr or aac audio.
  - Someone 4y ago
    We could bootstrap the format. First repeatedly send a p × q bit stream (with p and q prime) that teaches the aliens about a compressed format F1, then send a larger F1-encoded description of format F2, send a F2-encoded description of F3, …, send a Fn-encoded description of mp4, and then send a mp4-encoded video. For a simple version of that, see https://rosettaproject.org/ https://rosettaproject.org/, which uses large print to make it much more likely intelligent life will discover the small print. I don’t see video as high on the list of things we would like to send, though. There’s lots and lots of stuff we should send earlier, such as data on physical constants and chemical elements (things we expect any aliens to know, and that they could use to decipher our language)
    - gricardo99 4y ago
      on the list? how about close to pointless? i don’t see how bootstrapping the decoding gets you to anything other than another inscrutable format of data. without context about the point and use of the data i can’t imagine it would have any use. you need to know about the display devices and human audio, visual sensory and perception capabilities to make any use of the data. it would be like receiving some complex data format meant to generate a symphony of seismic waves that the observer enjoys through various specialized sensors, without knowing anything about this purpose of the data to begin with, let alone the many details specific to the use of the data by whatever devices “play” the data and how it’s perceived by the observer.
navi0 4y ago
…or Earth animal languages. We’ve managed to communicate with dolphins, but only to train them in the concepts we want, and they were smart enough to learn what we meant. But AFAIK, we still can’t understand any wild dolphin pod languages or dialects, let alone man’s best friend, elephants, other primates, or birds.
- flobosg 4y ago
  Only tangentially related, but your post reminded me of Ursula K. Le Guin’s “The Author of Acacia Seeds and Other Extracts from the Journal of Therolinguistics”: https://xenoflesh.files.wordpress.com/2017/03/ursula-k.-le-guin.pdf https://xenoflesh.files.wordpress.com/2017/03/ursula-k.-le-g...
- zeroth32 4y ago
  We understand dogs pretty well. It turns out their communication is not that complex.
- Asraelite 4y ago
  More like animal codes. Humans are the only species known to be biologically capable of language (expect for one controversial case of a chimp maaaaybe doing it). To be a language you need things like syntax and recursion.
  - kaba0 4y ago
    Why do you need syntax and recursion? From a very mathematical point of view, languages are just a set of words. And human language rules are not really grammatical rules for the most part - there are an endless number of exceptions.
    - Asraelite 4y ago
      > Why do you need syntax and recursion? Because that is the definition of a natural language. All human languages exhibit these features (except debatably Pirahã, but that's a weird case and is literally the only known example out of ~7000 languages). It's like the difference between a finite state machine and a Turing machine. For simple cases it may appear like they are equivalent, and with enough pattern matching and brute force you can always get by, but one will always be fundamentally more powerful than the other. This is the same with human language and animal communication. Animal brains are fundamentally incapable of grasping these concepts. > From a very mathematical point of view, languages are just a set of words. And human language rules are not really grammatical rules for the most part - there are an endless number of exceptions. I would say they're more like a complex partly logical partly statistical model than a set of words. Grammar is much, much more complicated than adjective-noun-verb. Once you create a good, comprehensive model of it for a language, true exceptions are very rare. Even accepting that there are expceptions, it doesn't make sense to say that just because humans don't use language with 100% predictability and rigor that it's equivalent to a random bag of words. This is true of programming and logical languages too. You need syntax which must be adhered to at least the vast majority of the time. Deviations have to be limited and understandable. Any sentence somebody says will be at least 95% known grammar with only a tiny bit of true innovation.
edgyquant 4y ago
That depends. If we can have back and forth with aliens it gets easier, or if they provide a Rosetta Stone (e.g. contact.) Otherwise it’ll be pretty hard if not impossible

sipior 4y ago

I think the only saving grace there is that the sentients sending such transmissions would presumably want to be understood, and would ensure a Rosetta Stone of sorts was packed along with the message; the common element of reference probably being the mechanism of the transmission medium itself. Contrast that with the Minoans here, who were just writing for other Minoans. But given how much culture and biology is inevitably encoded into even our human languages, one does sometimes wonder if we're even asking the right questions. I hope I live long enough to see one of the various SETI initiatives pan out into such an amazing discovery. As a child, I thought it inevitable; pushing fifty, I'm rather less sanguine.

showerst 4y ago

We _might_ share a set of universal constants with any other intelligence that they also care about. PI, the speed of light, binary, etc. In a weird sense specific to decoding transmissions we might share more of interest with them than Minoans, who were mostly likely recording the names of gods/heroes and accounting records for commodities that may no longer even exist.

ncmncm 4y ago
Tau, anyway. It seems unlikely anybody else would make the pi mistake.

Retric 4y ago

One big hope is it would include video to give context. Volume of text and consistency should also make decoding easier. Our Linear A samples are mostly fragments without context and they contain various inconsistencies.

jhauris 4y ago
I don't think sending video is a likely scenario if the sender wants to be understood. There are too many ways video could be encoded, even assuming the aliens have the same perceptual capabilities we have. Flipping it around, if we were going to do send a message to the universe would we send an h.265 bgr video with aac audio in streaming mp4 format? The only thing you can know for sure you have in common with someone receiving your radio signal is that they can receive radio signals, so using something about radio transmission would be a good bet for building a lexicon. Then there's the universal constants and mathematics which we could assume are known in common.
- Retric 4y ago
  Uncompressed video formats are vastly easier to decode than h.265 bgr or aac audio.
- Someone 4y ago
  We could bootstrap the format. First repeatedly send a p × q bit stream (with p and q prime) that teaches the aliens about a compressed format F1, then send a larger F1-encoded description of format F2, send a F2-encoded description of F3, …, send a Fn-encoded description of mp4, and then send a mp4-encoded video. For a simple version of that, see https://rosettaproject.org/ https://rosettaproject.org/, which uses large print to make it much more likely intelligent life will discover the small print. I don’t see video as high on the list of things we would like to send, though. There’s lots and lots of stuff we should send earlier, such as data on physical constants and chemical elements (things we expect any aliens to know, and that they could use to decipher our language)
  - gricardo99 4y ago
    on the list? how about close to pointless? i don’t see how bootstrapping the decoding gets you to anything other than another inscrutable format of data. without context about the point and use of the data i can’t imagine it would have any use. you need to know about the display devices and human audio, visual sensory and perception capabilities to make any use of the data. it would be like receiving some complex data format meant to generate a symphony of seismic waves that the observer enjoys through various specialized sensors, without knowing anything about this purpose of the data to begin with, let alone the many details specific to the use of the data by whatever devices “play” the data and how it’s perceived by the observer.

navi0 4y ago

…or Earth animal languages. We’ve managed to communicate with dolphins, but only to train them in the concepts we want, and they were smart enough to learn what we meant. But AFAIK, we still can’t understand any wild dolphin pod languages or dialects, let alone man’s best friend, elephants, other primates, or birds.

flobosg 4y ago
Only tangentially related, but your post reminded me of Ursula K. Le Guin’s “The Author of Acacia Seeds and Other Extracts from the Journal of Therolinguistics”: https://xenoflesh.files.wordpress.com/2017/03/ursula-k.-le-guin.pdf https://xenoflesh.files.wordpress.com/2017/03/ursula-k.-le-g...
zeroth32 4y ago
We understand dogs pretty well. It turns out their communication is not that complex.
Asraelite 4y ago
More like animal codes. Humans are the only species known to be biologically capable of language (expect for one controversial case of a chimp maaaaybe doing it). To be a language you need things like syntax and recursion.
- kaba0 4y ago
  Why do you need syntax and recursion? From a very mathematical point of view, languages are just a set of words. And human language rules are not really grammatical rules for the most part - there are an endless number of exceptions.
  - Asraelite 4y ago
    > Why do you need syntax and recursion? Because that is the definition of a natural language. All human languages exhibit these features (except debatably Pirahã, but that's a weird case and is literally the only known example out of ~7000 languages). It's like the difference between a finite state machine and a Turing machine. For simple cases it may appear like they are equivalent, and with enough pattern matching and brute force you can always get by, but one will always be fundamentally more powerful than the other. This is the same with human language and animal communication. Animal brains are fundamentally incapable of grasping these concepts. > From a very mathematical point of view, languages are just a set of words. And human language rules are not really grammatical rules for the most part - there are an endless number of exceptions. I would say they're more like a complex partly logical partly statistical model than a set of words. Grammar is much, much more complicated than adjective-noun-verb. Once you create a good, comprehensive model of it for a language, true exceptions are very rare. Even accepting that there are expceptions, it doesn't make sense to say that just because humans don't use language with 100% predictability and rigor that it's equivalent to a random bag of words. This is true of programming and logical languages too. You need syntax which must be adhered to at least the vast majority of the time. Deviations have to be limited and understandable. Any sentence somebody says will be at least 95% known grammar with only a tiny bit of true innovation.

edgyquant 4y ago

That depends. If we can have back and forth with aliens it gets easier, or if they provide a Rosetta Stone (e.g. contact.) Otherwise it’ll be pretty hard if not impossible