Loading ...
Sorry, an error occurred while loading the content.
 

Re: How much could one convey with a covert meta-language?

Expand Messages
  • Kelvin Jackson
    What about using different synonyms for a certain meaning to convey more info? Maybe when you use the native root the sentence is true, and when you use the
    Message 1 of 14 , Oct 31, 2009
      What about using different synonyms for a certain meaning to convey
      more info? Maybe when you use the "native" root the sentence is true,
      and when you use the latin-derived synonym, it is false. Or maybe you
      would use multiple different synonyms to convey coded text. Aka, the
      first in the list indicates a, the second b, etc (or probably
      something more sophisticated).

      On Oct 31, 2009, at 11:31 PM, Sai Emrys wrote:

      > A recent (and excellent) NYT article[0] made me think of this again.
      >
      > Namely: supposing that you have to communicate as much as possible but
      > are only given the ability to speak briefly in an L1 that is being
      > surveilled (and possibly coerced in detail), how much covert
      > information could you convey?
      >
      > I've thought of two possible channels so far.
      >
      > First, for non-tonal languages like English (let's assume for now that
      > that's the L1), you could use tones. But aside from trivial cases
      > (e.g. using broken tone on the primary verb of sentences that are
      > lies), I can't think of a good way to employ this. And doing it too
      > strongly would be noticeable.
      >
      > Second, you could refer to nonexistent codeworded relatives or the
      > like, or use some other explicitly prearranged codephrases. This would
      > require both a fair amount of memory, and a certain amount of freedom
      > of expression, which you might not have in all situations (e.g. if
      > you're being told at gunpoint to report in that everything's okay, or
      > if you've been given a script to read). However, it'd be better
      > preserved in case the communication isn't being recorded and is being
      > received without warning, since the person receiving the message (e.g.
      > a phone call) might not know how to understand a tonal system, but
      > would probably be able to remember a couple names.
      >
      > Ideas?
      >
      > - Sai
      >
      > [0] http://www.nytimes.com/2009/10/18/world/asia/18hostage.html?
      > pagewanted=all
    • Gary Shannon
      ... If the people doing the coercing/surveillance where not native speakers of your L1 (assume English for the moment) then it would be unlikely that they
      Message 2 of 14 , Nov 1, 2009
        On Sat, Oct 31, 2009 at 7:31 PM, Sai Emrys <sai@...> wrote:
        > A recent (and excellent) NYT article[0] made me think of this again.
        >
        > Namely: supposing that you have to communicate as much as possible but
        > are only given the ability to speak briefly in an L1 that is being
        > surveilled (and possibly coerced in detail), how much covert
        > information could you convey?

        If the people doing the coercing/surveillance where not native
        speakers of your L1 (assume English for the moment) then it would be
        unlikely that they would be able to distinguish between various
        accents, say Australian vs Bostonian vs Cockney vs Jamaican. With an
        agreed upon collection of a dozen different accents one would be able
        to communicate any one of a dozen different hidden meanings.

        --gary
      • Eldin Raigmore
        If you have a pre-arranged numeric code, you could reply in words of N syllables. For instance if you wanted to pass along three one four one five nine you
        Message 3 of 14 , Nov 1, 2009
          If you have a pre-arranged numeric code, you could reply in words of N
          syllables.
          For instance if you wanted to pass along "three one four one five nine"
          you could say something like
          "according to absolutely all communications psychopharmacologically ..."
        • Arnt Richard Johansen
          ... I m too lazy to find any references for this, but I seem to recall that stage magicians have for a long time used the initial consonants of words to
          Message 4 of 14 , Nov 1, 2009
            On Sat, Oct 31, 2009 at 08:31:18PM -0700, Sai Emrys wrote:
            > A recent (and excellent) NYT article[0] made me think of this again.
            >
            > Namely: supposing that you have to communicate as much as possible but
            > are only given the ability to speak briefly in an L1 that is being
            > surveilled (and possibly coerced in detail), how much covert
            > information could you convey?

            I'm too lazy to find any references for this, but I seem to recall that stage magicians have for a long time used the initial consonants of words to covertly convey information, usually numbers, to an assistant, which would be used for "mind reading" tricks.

            I think such a system would be very difficult to use, and result in unnatural sentences. (But magicians can get away with it, since their speech style is often melodramatic and circumlocutory.)

            --
            Arnt Richard Johansen http://arj.nvg.org/
            Read This Before Opening Package: According to Certain Suggested
            Versions of a Grand Unified Theory, the Primary Particles Constituting
            This Product May Decay to Nothingness Within the Next Four Hundred
            Million Years. --Susan Hewitt and Edward Subitzky
          • Gary Shannon
            ... As a former semi-profession magician (back in the early 1960 s) the method I m familiar with is a set of code phrases. An assistant goes out into the
            Message 5 of 14 , Nov 1, 2009
              On Sun, Nov 1, 2009 at 11:54 AM, Arnt Richard Johansen <arj@...> wrote:

              > I'm too lazy to find any references for this, but I seem to recall that stage magicians have for a long time used the initial consonants of words to covertly convey information, usually numbers, to an assistant, which would be used for "mind reading" tricks.
              >
              > I think such a system would be very difficult to use, and result in unnatural sentences. (But magicians can get away with it, since their speech style is often melodramatic and circumlocutory.)
              >

              As a former semi-profession magician (back in the early 1960's) the
              method I'm familiar with is a set of code phrases. An assistant goes
              out into the audience and asks for people to volunteer some object for
              the blindfolded magician to identify. The assistant asks different
              questions depending on what the object is. For example: "What am I
              holding now?" means a wrist watch; "Can you tell me what this is?"
              means a wallet; "Can you identify this object?" means a set of keys,
              and so on for a dozen or so common objects people might be carrying.

              --gary
            • Craig Daniel
              ... The best variant of two-person telepathy I m aware of (documented, IIRC, in Corinda) involves the magician explaining that he is going to have people take
              Message 6 of 14 , Nov 1, 2009
                On Sun, Nov 1, 2009 at 3:46 PM, Gary Shannon <fiziwig@...> wrote:
                > On Sun, Nov 1, 2009 at 11:54 AM, Arnt Richard Johansen <arj@...> wrote:
                >
                >> I'm too lazy to find any references for this, but I seem to recall that stage magicians have for a long time used the initial consonants of words to covertly convey information, usually numbers, to an assistant, which would be used for "mind reading" tricks.
                >>
                >> I think such a system would be very difficult to use, and result in unnatural sentences. (But magicians can get away with it, since their speech style is often melodramatic and circumlocutory.)
                >>
                >
                > As a former semi-profession magician (back in the early 1960's) the
                > method I'm familiar with is a set of code phrases. An assistant goes
                > out into the audience and asks for people to volunteer some object for
                > the blindfolded magician to identify. The assistant asks different
                > questions depending on what the object is. For example: "What am I
                > holding now?" means a wrist watch; "Can you tell me what this is?"
                > means a wallet; "Can you identify this object?" means a set of keys,
                > and so on for a dozen or so common objects people might be carrying.

                The best variant of two-person telepathy I'm aware of (documented,
                IIRC, in Corinda) involves the magician explaining that he is going to
                have people take out objects from their pockets and hold them up, and
                he is going to go around and point at various people and ask his
                assistant (who is not only blindfolded but also facing away from the
                audience) to call out what their objects are. As he walks through the
                crowd, he points at volunteers and says "What's this?" At which point
                the assistant names the item, and the magician says "Yes" to confirm.
                He repeats this process several times, always using a very neutral
                tone of voice and no phrasing ever changes.

                Like Gary's method, it relies on a pre-arranged list of common
                objects; the principle probably couldn't be extended to more than
                about a dozen possible messages without including other techniques as
                well, but it seems like a gold standard for the degree of covertness.

                (I would dearly love to actually build up the skill required to
                perform this effect. Naturally it requires a great deal of practice
                with the same assistant you're going to be performing with.)
              • Eric Christopherson
                ... So how *does* the assistant know?
                Message 7 of 14 , Nov 1, 2009
                  On Nov 1, 2009, at 4:51 PM, Craig Daniel wrote:

                  > On Sun, Nov 1, 2009 at 3:46 PM, Gary Shannon <fiziwig@...>
                  > wrote:
                  >> On Sun, Nov 1, 2009 at 11:54 AM, Arnt Richard Johansen
                  >> <arj@...> wrote:
                  >>
                  >>> I'm too lazy to find any references for this, but I seem to recall
                  >>> that stage magicians have for a long time used the initial
                  >>> consonants of words to covertly convey information, usually
                  >>> numbers, to an assistant, which would be used for "mind reading"
                  >>> tricks.
                  >>>
                  >>> I think such a system would be very difficult to use, and result
                  >>> in unnatural sentences. (But magicians can get away with it, since
                  >>> their speech style is often melodramatic and circumlocutory.)
                  >>>
                  >>
                  >> As a former semi-profession magician (back in the early 1960's) the
                  >> method I'm familiar with is a set of code phrases. An assistant goes
                  >> out into the audience and asks for people to volunteer some object
                  >> for
                  >> the blindfolded magician to identify. The assistant asks different
                  >> questions depending on what the object is. For example: "What am I
                  >> holding now?" means a wrist watch; "Can you tell me what this is?"
                  >> means a wallet; "Can you identify this object?" means a set of keys,
                  >> and so on for a dozen or so common objects people might be carrying.
                  >
                  > The best variant of two-person telepathy I'm aware of (documented,
                  > IIRC, in Corinda) involves the magician explaining that he is going to
                  > have people take out objects from their pockets and hold them up, and
                  > he is going to go around and point at various people and ask his
                  > assistant (who is not only blindfolded but also facing away from the
                  > audience) to call out what their objects are. As he walks through the
                  > crowd, he points at volunteers and says "What's this?" At which point
                  > the assistant names the item, and the magician says "Yes" to confirm.
                  > He repeats this process several times, always using a very neutral
                  > tone of voice and no phrasing ever changes.

                  So how *does* the assistant know?

                  >
                  > Like Gary's method, it relies on a pre-arranged list of common
                  > objects; the principle probably couldn't be extended to more than
                  > about a dozen possible messages without including other techniques as
                  > well, but it seems like a gold standard for the degree of covertness.
                  >
                  > (I would dearly love to actually build up the skill required to
                  > perform this effect. Naturally it requires a great deal of practice
                  > with the same assistant you're going to be performing with.)
                • Craig Daniel
                  ... Hah, that somehow got deleted as I was editing my message. The information is conveyed through the duration of the pause between confirming one object and
                  Message 8 of 14 , Nov 1, 2009
                    On Sun, Nov 1, 2009 at 6:27 PM, Eric Christopherson <rakko@...> wrote:
                    > So how *does* the assistant know?

                    Hah, that somehow got deleted as I was editing my message.

                    The information is conveyed through the duration of the pause between
                    confirming one object and asking about the next.
                  • Patrick Dunn
                    ... Just as a guess, maybe there s a pre-memorized sequence of items. Say, keys, wallet, coin, watch. The magician just needs to point to objects in that
                    Message 9 of 14 , Nov 1, 2009
                      On Sun, Nov 1, 2009 at 5:27 PM, Eric Christopherson <rakko@...> wrote:
                      > On Nov 1, 2009, at 4:51 PM, Craig Daniel wrote:
                      >
                      >> On Sun, Nov 1, 2009 at 3:46 PM, Gary Shannon <fiziwig@...> wrote:
                      >>>
                      >>> On Sun, Nov 1, 2009 at 11:54 AM, Arnt Richard Johansen <arj@...>
                      >>> wrote:
                      >>>
                      >>>> I'm too lazy to find any references for this, but I seem to recall that
                      >>>> stage magicians have for a long time used the initial consonants of words to
                      >>>> covertly convey information, usually numbers, to an assistant, which would
                      >>>> be used for "mind reading" tricks.
                      >>>>
                      >>>> I think such a system would be very difficult to use, and result in
                      >>>> unnatural sentences. (But magicians can get away with it, since their speech
                      >>>> style is often melodramatic and circumlocutory.)
                      >>>>
                      >>>
                      >>> As a former semi-profession magician (back in the early 1960's) the
                      >>> method I'm familiar with is a set of code phrases. An assistant goes
                      >>> out into the audience and asks for people to volunteer some object for
                      >>> the blindfolded magician to identify. The assistant asks different
                      >>> questions depending on what the object is. For example: "What am I
                      >>> holding now?" means a wrist watch; "Can you tell me what this is?"
                      >>> means a wallet; "Can you identify this object?" means a set of keys,
                      >>> and so on for a dozen or so common objects people might be carrying.
                      >>
                      >> The best variant of two-person telepathy I'm aware of (documented,
                      >> IIRC, in Corinda) involves the magician explaining that he is going to
                      >> have people take out objects from their pockets and hold them up, and
                      >> he is going to go around and point at various people and ask his
                      >> assistant (who is not only blindfolded but also facing away from the
                      >> audience) to call out what their objects are. As he walks through the
                      >> crowd, he points at volunteers and says "What's this?" At which point
                      >> the assistant names the item, and the magician says "Yes" to confirm.
                      >> He repeats this process several times, always using a very neutral
                      >> tone of voice and no phrasing ever changes.
                      >
                      > So how *does* the assistant know?
                      >

                      Just as a guess, maybe there's a pre-memorized sequence of items.
                      Say, "keys, wallet, coin, watch." The magician just needs to point to
                      objects in that order. I suppose the tricky bit would be if no one in
                      the audience holds up one of the prearranged items.
                    • Larry Sulky
                      Next time I go to see a magic show I m going to have, in my pocket, a rubber chicken, a grapefruit spoon, a doily, and a gas mask.
                      Message 10 of 14 , Nov 1, 2009
                        Next time I go to see a magic show I'm going to have, in my pocket, a rubber
                        chicken, a grapefruit spoon, a doily, and a gas mask.
                      • David E
                        I ll bring extras to pass out to the other patrons.
                        Message 11 of 14 , Nov 1, 2009
                          I'll bring extras to pass out to the other patrons.

                          On Sun, Nov 1, 2009 at 6:54 PM, Larry Sulky <larrysulky@...> wrote:

                          > Next time I go to see a magic show I'm going to have, in my pocket, a
                          > rubber
                          > chicken, a grapefruit spoon, a doily, and a gas mask.
                          >
                        • Peter Bleackley
                          ... Kennings. Have a system of non-obvious but memorable metonyms for some of the things you want to discuss (eg the Bread s Voice for a priest). Used a bit in
                          Message 12 of 14 , Nov 2, 2009
                            staving Sai Emrys:
                            > A recent (and excellent) NYT article[0] made me think of this again.
                            >
                            > Namely: supposing that you have to communicate as much as possible but
                            > are only given the ability to speak briefly in an L1 that is being
                            > surveilled (and possibly coerced in detail), how much covert
                            > information could you convey?
                            >
                            > I've thought of two possible channels so far.
                            >

                            > Ideas?
                            >

                            Kennings. Have a system of non-obvious but memorable metonyms for some
                            of the things you want to discuss (eg the Bread's Voice for a priest).
                            Used a bit in ancient germanic verse, and similar to what was done by
                            Navaho code talkers.

                            Pete
                          • Eldin Raigmore
                            Another method I got from an Asimov story: Each word represents a Morse-code letter. Every counts as a dash and every counts as a dot. Other letters
                            Message 13 of 14 , Nov 3, 2009
                              Another method I got from an Asimov story:
                              Each word represents a Morse-code letter.
                              Every <t> counts as a dash and every <i> counts as a dot.
                              Other letters don't count.
                              So "titrate is inimical" would be -.-- . ... which would be Y E S .
                            Your message has been successfully submitted and would be delivered to recipients shortly.