Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A deep code found in Benjamin Franklin's 13 virtues

  1. Dec 3, 2006 #1
    A deep code found in Benjamin Franklin's 13 virtues

    Graphic representation below:

    Discovered after total binge eating, and after hitting a new high of 173 pounds (is me). Sorry my stomach =P

    [​IMG]

    [​IMG]

    Two picture sources incase if one of them disappears.
     

    Attached Files:

  2. jcsd
  3. Dec 3, 2006 #2
    Sure why not. The presumption I began with was to find hidden meaning by rearranging the lines so that they contain more information. This was a tough task, but the solution was simple. Number each virtue by it's position in the list, from 1 to 13. Then categorize them into groups where the sum of each group (by the numbers of its virtues) is equal. Since the numbers 1 to 13 sum up to 91, this means that there needs to be 7 groups totalling 13, because by factoring 91 we get the prime numbers 7 and 13 (7 groups all of value 13). So matching was not random or intuitive, it was the necessary result of this repeatable, verifiable procedure.

    With great luck, the virtues happen to match each other. Even more, I got down to the bottom lines in the image, which show just how powerful simplicity can be:

    [​IMG]

    [​IMG]
     
  4. Dec 3, 2006 #3

    DaveC426913

    User Avatar
    Gold Member

    Maybe you should post your source material and then an explanation of what you'rfe doing by the transform.

    Or maybe you should get out of the kitchen and exercise a little more.
     
  5. Dec 3, 2006 #4

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    So you started by assuming there was "hidden meaning" and you found it! Is anyone surprised? Do you understand why that is a scientifically unacceptable procedure?
     
  6. Dec 3, 2006 #5
    I did just that. Raking and brooming leaves outside ;). I plan not to eat but drink today as the saliva I have is supposed to supress my appetite (i think).

    The list I got was from the Agon Shu (http://www.agon.org/us/DharmaTalks/DharmaTalks01_01.html), the descriptions of which have less of ol' dialect. They are in the same order however as the ones on (http://www.school-for-champions.com/character/franklin_virtues.htm) and (http://en.wikipedia.org/wiki/Benjamin_Franklin):

    Step one, number the virtues from 1 to 13

    To categorize, you need groups. To minimize arbitration, make a simple rule that determines which terms are to be matched. Since this must be done quantiatively and not qualitatively, so as to allow for only one solution, we must use numbers. If we make groups, is it arbitrary to distribute them randomly. Therefore there need to be a single rule that follows which will predetermine the groups themselves. The sum of numbers 1 to 13 happen to be equal to a multiple of 13, or 13*7. Multiplication will not work because 1,3,5,7,11,13 are prime numbers and is it therefore impossible to categorize with multiplication. For the same reason, we cannot use division. Subtraction cannot work because then one group will have to have either two minuends or two subtrahends. Therefore we use addition. To assure that there is only one way, there has to be a rule that allows for only one solution. The rule:

    "The 7 groups assigned to must result in a sum of 13 for each group's virtues' list numbers."

    This means that virtue 1 must go with virtue 12
    This means that virtue 2 must go with virtue 11
    This means that virtue 3 must go with virtue 10
    This means that virtue 4 must go with virtue 9
    This means that virtue 5 must go with virtue 8
    This means that virtue 6 must go with virtue 7

    That there is not a virtue going with itself is something which cannot remove the validity of the rule.

    We are left with the following groups:

    The groups are now:
    This, in fact ends, the part where you can get one result only. So I was wrong when I implied that there was only one solution at the very end...

    How do these groups relate to each other? A lot could be said about how to oragnize them. But once groups become larger than they need to be, they become obfuscated. It is not simple to merge these groups even more so that there is 4 items in the majority of groups. Assuming that endless grouping is meaningless without emphasis, we decide to indent lines. Reordering them a second time would be as if the previous reordering was a mistake. We could try:

    But this adds nothing new that is dynamic, like adding constant to an integration. We should rather have:

    or

    Or any other equivalent. Now the interpretation part begins. Interpretation is colored by life experiences and therefore lacks pure objectivity. Unlike the first section, mathematics is not involved. However, we may associate the virtues according to a logical structure such as "A, or else, not B". The task is then to find what can be A and B. Here, the virtues are indented in correspondence to their groups (per above):

    Code (Text):

    A    B
    1&12 13
    2    7
    11   6
    3    8
    10   5
    4    9
     

    Some may come and disagree with it, or perhaps find an excess of relationships, and that is a flaw of my design. But I believe that even if the process itself didn't make sense, it allowed me to reach conclusions (that make sense to me) that would otherwise be harder for me to imagine.
     
  7. Dec 3, 2006 #6
    I categorize it this way. If am looking at a state in order to deduce other states, it's not science. Therefore, what I did was not science.

    The previous paragraph was an oversimplification. The way I spell it out is this... If you observe a state, say a tv missing some of its buttons (i.e. a "missing some of its buttons" tv), you cannot deduce any other states (by knowing only that, you don't know if it is in space, what color is it, if it has a remote control, whether it has a infrared sensor in place or not, etc., or even whether it has a frame (because it could a hologram or even a flat picture in a book)). However, if you are observing an activity (such as the onscreen results occuring after a particle accelerator event, the event where a by product emerges from mixing two chemicals, or the movement of two colliding masses), then you can deduce a state (or states).

    Scientists must at times observe the animate, when the experiment is running (without this method we could not measure any speed or acceleration, or determine the stimulus which generates an increase in temperature and entropy). But what I was analyzing in this post was not moving. I was not presenting myself a physical event but a group of attributions which I manipulated. Veritably, what I provided did not lead to any scientific discoveries.
     
    Last edited: Dec 3, 2006
  8. Dec 3, 2006 #7
    i dont get it:uhh:
     
  9. Dec 3, 2006 #8
    All things that are observed are observed in a non-zero period of time.

    An observation that began and ended at the same time did not occur.

    States are defined at the moment.

    In science, we observe events which are in reality many different states in sequence.

    Because of our observation of events, we can know the states of each "frame" of the event.

    Without observation of events, we may observe a "frame" of an event, but will have trouble determining that event correctly without a theory based on a similar event from which to determine what the event is.

    This "frame" of an event, by itself, is not animate. It is a still, containing far less information than 1000 "frames" of similar detail would.

    So what? This is what.

    I did not present any material evidence. Even then, material evidence is only evidence because it is associated with a known event. However, there is no evidence for a secret code by Benjamin Franklin that has any resemblance to what I presented. What I presented or talked about was a collection of attributes not an event (or events). Non-events are not proof of anything.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: A deep code found in Benjamin Franklin's 13 virtues
  1. Deep Thoughts (Replies: 48)

  2. Deep proverbs. (Replies: 6)

Loading...