Randomness in Tarot

I'm awful at math and statistics but I can say that since May of this year I have done 6 reads along the same theme regarding personal growth. I used 5 different decks - the last two were the same deck but single card draws. No spread was more than 4 cards.

I have drawn the 9 of S 3 times
10 of S 3 times
Temperance twice
Hierophant twice

Those were my duplicates

There were a total of 13 cards drawn.

All of the cards were drawn go-fish style in a pool on a table.

Just saying that even if it's random it's highly coincidental .

I personally don't care if a tarot draw is random or purposeful at all. What matters is what that card invokes in you when you look at it. Life is a series of emotional and psychological reactions in the person. That's what creates experience. It's not what happens but how you interpret it that makes for an experience.

I personally have no interest in trying to prove or disprove the relevance/validity/scientific evidence for something like the occult. It's like trying to prove the validity of fairytales. why bother? The purpose of fairytales is to add colour and enchantment to life. That's exactly how I feel about the occult too. I hate the idea that my life can be reduced to a series of random, empirically quantifiable and proven events. Where's the fun in that? Life is serious and boring enough without using science to bludgeon every last scrap of enchantment from it.

I tend to see skeptics as party-poopers extraordinaire.

SwordOfTruth said:

I personally don't care if a tarot draw is random or purposeful at all. What matters is what that card invokes in you when you look at it. Life is a series of emotional and psychological reactions in the person. That's what creates experience. It's not what happens but how you interpret it that makes for an experience.

I personally have no interest in trying to prove or disprove the relevance/validity/scientific evidence for something like the occult. It's like trying to prove the validity of fairytales. why bother? The purpose of fairytales is to add colour and enchantment to life. That's exactly how I feel about the occult too. I hate the idea that my life can be reduced to a series of random, empirically quantifiable and proven events. Where's the fun in that? Life is serious and boring enough without using science to bludgeon every last scrap of enchantment from it.

I tend to see skeptics as party-poopers extraordinaire.

Click to expand...

hahaha I'm with you SwordOfTruth! I like tracking my readings just to see how accurate they are and I'm also intrigued by the possible entanglement of science and mystic. I don't think it's about reducing either. But I do agree with you that not everything is or should be turned into statistics and technical details. We are not robots and we don't operate on binary code. I love your definition of life: a series of emotional and psychological reactions in a person. Indeed.

I once read this definition that really stuck: Life is enjoying and suffering existence.
And tarot can help make sense of this existence

chaosbloom said:

I wanted to start a thread exactly on this topic a few days ago but got busy with a few other things. Finding that a thread was opened on the same topic but denying nonrandomness is some pretty funny synchronicity. Anyway, there's been some precedence on this.

http://www.eheart.com/BOOKS/fingers/tarot-jane.pdf

It's pretty easy to reproduce the experiments, we could do that collectively by compiling raw data of frequencies of cards appearing in spreads over the course of a month or any other time frame. In the above article, there's also the idea that without intent, the deck approaches true randomness while intent decreases randomness.

As for quick and dirty explanations for the nonrandomness, repetitive ritualistic habits in shuffling and cutting might affect card distribution. But that's easy to control and the article above claims that a control group with normal playing cards which I assume used the same shuffling and cutting techniques didn't show the same kind of randomness. I'm not going to decide either way until I examine my own data at some point but I've seen recurrent cards and card themes so many times that I have little doubt about nonrandomness.

Click to expand...

What a fantastic article! Thank you so much for sharing it with us (me), chaosbloom. Her results are really quite compelling, as is the replication of her experiment (with equivalent results) by another researcher. And while someone else might be able to poke holes in her method, I certainly can't; it's very similar to what I was hoping to achieve with my own experiment, but was conducted on a much larger (and therefore more significant) scale. I'm really, really glad you shared this article. It's enough in the way of real, quantifiable empirical evidence to give me pause and make me think very seriously about my views.

earthair said:

I think your initial experiment needs tweaking a bit JackofWands!

I suggest that those of us who are interested all draw a 3 card spread for a serious question concerning the future, maybe about a global issue , then without telling anyone else, we tell JackofWands (plus one other person who acts as independent moderator) which cards we drew via Private Msg and he collates the data?

We also need two control groups... 1- who don't ask any question and 2- who ask any question they like.

Click to expand...

I would love to participate in a large-scale experiment like this! If enough other people would be interested, please do let me know. I would gladly compile all of the data (Microsoft Excel and I are very good friends) and would share all the results--along with all the formulae I used, so that others could verify that I didn't make some grievous error--here on the forum.

In order for a chi-square test to be reliable, the expected value (e in the study chaosbloom linked to) for each possibility must be at least 5. This means that an undertaking like this would have to have 78x5 = 390 cards drawn in total. If we were doing three-card readings, then we would need a total of 130 readings total conducted, each (ideally) on the same question, using the same spread, and conducted with intent.

Realistically, I don't know if it would be feasible to gather 130 ATers together for a project like this, especially if you want to have a control group or even two. (Strictly speaking, a control group mightn't be necessary with a chi-square test, because we're testing for deviation from a random distribution; the model of pure randomness acts as our control. Plus, we're not exactly looking to get this study published in a peer-reviewed journal, although any aspiring Ph.D. candidates looking for thesis topics might do well to take note of our work here.)

Another alternative would be to do a larger spread, which would require fewer people. Six cards per reading would only necessitate 65 participants, and 10 cards per reading (say, a Celtic Cross) would only require 39. This would still be extremely ambitious, but is more realistic than trying to gather more than a hundred readers together for something that many consider fundamentally inconsistent with the nature of Tarot.

Mods, any thoughts on this? If we were to really seriously undertake this project, do you know of a way to make it feasible?

One possibility might be to set up something like a reading circle (assuming there was genuine interest in undertaking a project like this and assuming it was kosher with the moderators). We could post one question and a spread for everyone to use, and then people could post their draws (and interpretations, if they chose) in reply to the thread. Afterwards, the circle host would have all of the relevant information gathered in one place, and would be able to perform any and all necessary analysis.

Thoughts? Dealing with the selection of a question would come later, after we hammered out the practicality of actually making an undertaking like this happen. That said, I think we would have to be very careful with the choice of question, because so many readers have different beliefs about what kinds of question "work" or even just about which questions are ethical. (For example, some readers strictly avoid predictive readings.)

SwordOfTruth said:

I personally don't care if a tarot draw is random or purposeful at all. What matters is what that card invokes in you when you look at it. Life is a series of emotional and psychological reactions in the person. That's what creates experience. It's not what happens but how you interpret it that makes for an experience.

I personally have no interest in trying to prove or disprove the relevance/validity/scientific evidence for something like the occult. It's like trying to prove the validity of fairytales. why bother? The purpose of fairytales is to add colour and enchantment to life. That's exactly how I feel about the occult too. I hate the idea that my life can be reduced to a series of random, empirically quantifiable and proven events. Where's the fun in that? Life is serious and boring enough without using science to bludgeon every last scrap of enchantment from it.

I tend to see skeptics as party-poopers extraordinaire.

Click to expand...

At least we're extraordinaire, then!

I can understand the desire to leave off skepticism with some things--does it really matter, after all?--but I would say that for me, it's not so much a matter of trying to bludgeon enchantment from the world as it is a question of simple curiosity. I'm curious about the world and want to know more about it. I'm interested in Tarot and want to understand it to the greatest extent that I can. And for me, at least, the first step to real understanding of anything is a strong dose of skepticism.

"The Pyrrhonian skeptic, Sextus Empiricus, questioned the apodicticity of inductive reasoning because a universal rule cannot be established from an incomplete set of particular instances: "When they propose to establish the universal from the particulars by means of induction, they will effect this by a review of either all or some of the particulars.But if they review some, the induction will be insecure, since some of the particulars omitted in the induction may contravene the universal; while if they are to review all, they will be toiling at the impossible, since the particulars are infinite and indefinite."

I am not sure how far your research will get you. The phrase "the particulars are infinite and indefinite" specifically strikes me as describing the tarot.

JackofWands said:

Thoughts? Dealing with the selection of a question would come later, after we hammered out the practicality of actually making an undertaking like this happen. That said, I think we would have to be very careful with the choice of question, because so many readers have different beliefs about what kinds of question "work" or even just about which questions are ethical. (For example, some readers strictly avoid predictive readings.)

Click to expand...

You're welcome about the article, I found it a few days ago as I was getting things together to quantify my recorded spreads so far. Here are some thoughts.

First of all, we're talking about magic. Think three cards, five cards, seven cards. Not six. But jokes aside. After we determine if its feasible to get enough data, the point of what hypothesis we're actually testing is more complicated than it looks. Sure, intent seems to affect randomness and we can determine that but we should definitely avoid selecting a loaded question. There's the widespread belief that either sometimes or always it is the emotional/mental state of the user that affects randomness and not an outside factor. Selecting a yes or no question for example that half the test subjects influence towards a no and the other half influence for a yes would skew the results.

This area is very close to the untestable because of the myriad unknown factors involved but with enough caution and small steps we can actually go far.

Padma said:

"The Pyrrhonian skeptic, Sextus Empiricus, questioned the apodicticity of inductive reasoning because a universal rule cannot be established from an incomplete set of particular instances: "When they propose to establish the universal from the particulars by means of induction, they will effect this by a review of either all or some of the particulars.But if they review some, the induction will be insecure, since some of the particulars omitted in the induction may contravene the universal; while if they are to review all, they will be toiling at the impossible, since the particulars are infinite and indefinite."

I am not sure how far your research will get you. The phrase "the particulars are infinite and indefinite" specifically strikes me as describing the tarot.

Click to expand...

This is an excellent point. I think that no research--and in particular, nothing conducted on such a small scale as what I have the resources to orchestrate--could definitively prove or disprove a certain set of views about Tarot. The particulars are, as you note, infinite. But that does not make reasoning by induction completely invalid.

We can set forth a basic set of assumptions: if Tarot functions according to principle A, then we would expect it to look like B. And then we can look to see whether external reality is consistent with those assumptions. If we find that it is, we haven't proven anything, but we can continue about our daily business with at least a little bit more support for our worldview. And if we find that it's not, then we call into question either principle A or the assumption that A is sufficient for B.

In the case of my initial experiment, the null hypothesis I was working with was that Tarot was random (because I think this is the default claim; I'd argue that the burden of proof is on anyone who would claim otherwise). If Tarot is random, I would expect 100 daily draws to turn up a random-looking distribution, and this is exactly what happened in my case. That does not definitively prove that my point of view is correct, but it at least means that my opinion is consistent with externally observed reality.

The analysis cited by chaosbloom is interesting to me, because it presents a set of observations that are not consistent with my initial assumption. This does not necessarily mean that Tarot isn't random, any more than my experiment means that it is, but it's an observation that needs to be taken into consideration if I want my opinions to be grounded in the empirical world. And for me, it invites further research. If I can somehow replicate the results of that experiment, then it gains serious credibility. If it can't be replicated, then its legitimacy is questionable and I'll be inclined to maintain my original point of view.

None of this is a matter of definitive proof. It's merely about trying to make my understanding of the world as consistent with external reality as I possibly can.

chaosbloom said:

You're welcome about the article, I found it a few days ago as I was getting things together to quantify my recorded spreads so far. Here are some thoughts.

First of all, we're talking about magic. Think three cards, five cards, seven cards. Not six. But jokes aside. After we determine if its feasible to get enough data, the point of what hypothesis we're actually testing is more complicated than it looks. Sure, intent seems to affect randomness and we can determine that but we should definitely avoid selecting a loaded question. There's the widespread belief that either sometimes or always it is the emotional/mental state of the user that affects randomness and not an outside factor. Selecting a yes or no question for example that half the test subjects influence towards a no and the other half influence for a yes would skew the results.

This area is very close to the untestable because of the myriad unknown factors involved but with enough caution and small steps we can actually go far.

Click to expand...

Three cards is doable. Five cards would require 78 different readings. Seven doesn't divide evenly, but would require about 56 different people performing readings. Personally, I think that even 39 is far more than we would realistically be able to get involved in a project like this (although it's a pity, because I really would be interested to see the outcome). Truth be told, I don't expect much to come of this idle planning. But who knows?

Like I said, I think we should set aside the question of questions for the moment. Of course yes-or-no questions would probably not be the best idea, although for me (still finding the idea of an individual's will magically affecting the outside world dubious, at best) that's more a matter of the types of question that are most easily read for, rather than of skewing the results.

Yes, there are lots of potential variables, but once again, we're not necessarily looking for definitive, irrefutable proof one way or another (or at least I'm not). I'd say that this is more a matter of controlling the variables that we can and looking at what patterns arise, if any.

One thing to consider, which crossed my mind when I read the article you linked but which I don't have the mathematical prowess to address on my own. When you do a multiple-card reading, the odds of drawing a given card change with each draw. With the first, you have a 1/78 chance, but with the second, you have a 1/77, and with the third, a 1/76. For the mathematicians here (I know there are a couple): wouldn't this affect the chi-square values for each card? Would it not be necessary to calculate multiple parallel chi-squares? If so, how would this affect the number of readings we would need to conduct? (i.e. Would it be necessary to perform more than 390 readings total in order to have sufficient data?) And is there any chance (asked the skeptic) that this question of varying degrees of freedom would have positively skewed the calculations in the analysis chaosbloom linked?

Unfortunately, my mathematical skill is not sufficient to answer these questions, and I may even be wrong in posing them. The woman who conducted this study has a Ph.D. in physics, so I'm inclined to think that she knows her stuff and I'm just being ignorant. But hopefully someone here on the forum can straighten me out in this regard.

JackofWands said:

In the case of my initial experiment, the null hypothesis I was working with was that Tarot was random (because I think this is the default claim; I'd argue that the burden of proof is on anyone who would claim otherwise). If Tarot is random, I would expect 100 daily draws to turn up a random-looking distribution, and this is exactly what happened in my case. That does not definitively prove that my point of view is correct, but it at least means that my opinion is consistent with externally observed reality.

Click to expand...

Knowing tarot as I do, it seems to me the cards were playing a kind of practical joke on you...If tarot reflects our mindset and expectations, it was duly reflecting that back to you! You got the answer you were intending. The tarot saw your question of randomness, and proceeded to divine it for you...rather accurately, it appears.

I have often noted the dry, tongue-in-cheek wit of the cards...

I see scientists as really bad dog trainers... Trying to rub your nose in it.....

Padma said:

Knowing tarot as I do, it seems to me the cards were playing a kind of practical joke on you...If tarot reflects our mindset and expectations, it was duly reflecting that back to you! You got the answer you were intending. The tarot saw your question of randomness, and proceeded to divine it for you...rather accurately, it appears.

I have often noted the dry, tongue-in-cheek wit of the cards...

Click to expand...

I'd welcome being proved wrong! But personally, I don't find the argument of "you got the answer you were intending" satisfying. It presupposes a certain nature to the cards, and then reinterprets external events to fit with that supposition. It's essentially saying, "The cards are magic, and if you have evidence that suggests otherwise, that's only because the cards are magic". Sort of like the old medieval gag of "You're a crop-blighting witch, and if you claim not to be, that's because crop-blighting witches lie". It doesn't leave much room for other possibilities.

I'm open to changing my views, and I find something like chaosbloom's study very thought-provoking. That's especially why I'm interested to see if it can be replicated (by more spiritual readers than myself, lest my rationalism contaminate the results). But as far as my personal views go, I do need evidence.