Tarot Carrot

[nonbeliever]

Forgive me if I'm re-re-asking an old question; I'm new here.

Is there a reasonable way to test general claims of successful tarot reading? Have any tarot challenges been conducted at JREF in the past?

 

[drkitten]

Forgive me if I'm re-re-asking an old question; I'm new here.

Is there a reasonable way to test general claims of successful tarot reading? Have any tarot challenges been conducted at JREF in the past?

I'm not aware of any tests that have been done.

But there's no reason in theory that a test could not be done. The hard part would be to get the tarot practitioner to detail exactly what she can and can't do in a form that would be objectively validatable. Since the claims of tarot practitioners are sufficiently varied, there's no one-size-fits-all test.

But one obvious possibility -- backed by antiquity -- would be divination of the sex of an expected child. Present the reader with twenty pregnant women and have the diviner determine whether she's carrying a boy or a girl, results to be confirmed via standard medical means or after delivery. You might need to plug a loophole regarding multiple births, and of course would need to find the necessary women. But the test itself is easily runnable.

What else do tarot practitioners claim to do? I've seen some that will tell you whether or not you'll get a promotion soon -- again, easy enough to check if you get the right group. I've seen some that will tell you whether or not you will get married soon. Some will tell you about your state of health, and so forth. Any of the "standard" divination tests would work if the practitioner can be convinced to state anything forthrightly.

 

[thomps1d]

This isn't an official answer, but this is how I would see it:

The problem with tarot readings is that, like fortune cookies or horoscopes, they thrive on being intentionally vague.

Perhaps a test like this could work:

Two decks of tarot cards are present. The test subject shuffles and cuts one deck, then leaves the room. One of the two decks is picked randomly (coin flip, computer program, picked out of a hat, etc), and then the other deck is put away. You and the test subject enter the room (having had no contact with each other previously). You are presented with the selected deck of cards; you have no way of knowing if it was shuffled by the test subject or pre-shuffled by the testers. You deal the cards and do the reading - without asking the test subject questions, you tell him/her what the cards say. You leave the room, and the test subject evaluates you on accuracy of verifiable facts that you "read".

ETA: I guess I'm too slow. drkitten beat me to the punch, and had some good ideas.

 

[GzuzKryzt]

I agree, drkitten's proposed divination of the sex of an expected child does not only prove a valid test of divining powers, it is easily testable and as easily verifiable - thanks to modern science.

A tarot reader woorth his salt should at least get 16 out of 20 correct to start convincing us, shouldn't he? (See Carina Landin protocol.)

Nonbeliever, bring on the applicants.

 

[chillzero]

Tarot reading was going to be my big entry to the MDC when I first arrived here. I soon realised it would be incredibly difficult to determine a protocol.

Tarot readings are based on the 'belief' that you are laying out a path for the future, and that there are still endless options. All you do by reading someone's card is throw a little light on one part of a map, identifying one route they might take.

Should your reading turn out to be way off, then clearly they took a different decision somewhere along the way - a right turn onto a side lane where you had lit up the left turn to the motorway, so to speak.

Even the act of giving / receiving the tarot reading could alter the path ahead. One event that I used to believe was proof of the effectiveness of tarot is quite embarrassing. I read for myself before I went to visit someone, and although I had been really looking forward to the trip, the cards spelt doom and disaster - chaos, arguments and misunderstandings. At first I was confused - how could the cards be so wrong? Then I sat and had a meditation on the journey ahead. I realised that I harboured some anger at the person for being so far away when I needed them with me, and necessitating the trip in the first place. I was upset about their actions, and no doubt this would have come across, particularly after the long journey I had to make to get there. I would be tired, irritable, and no doubt ready for an arguement. I calmed myself, thought through my reasons for agreeing to go on the trip, and reassessed the situation. I decided that I needed to go ahead with the trip, but not dragging any of the emotional baggage with me.

I re-did my reading, and now the cards looked great. I felt reassured by this, and went on the trip with more confidence, and a lighter heart that I had avoided a disaster. Of course, had the cards still spelled doom, the story may have ended differently, I don't know - maybe I would have decided that the trip was a bad idea and cancelled. In any case, I think that it is clear that I fitted the reading around things I actually knew. The trip went well, but only because I chose to ignore warning signs about the relationship while I was there.

Card readings are too fluid, and not designed to give black and white yes/no answers, so a protocol will be incredibly difficult to settle.

 

[simonmaal]

The problem with testing tarot readings is that statements by the reader can be interpreted by the participant in so many different ways. Meanings, causes and relationships between events can be extrapolated from the vagaries of everyday life, for no reason other than the basic human need to create order amidst the chaos of the universe. In short, the participant imposes his or her personal meanings onto extremely vague statements, and this is especially true with people who are superstitious, paranoid or manic (precisely the kind of person who would seek a reading in the first place).

The biggest problem with tarot cards is, of course, that they rely heavily on suggestion, psychological trickery and cold-reading. I just started a thread on this subject:

http://www.internationalskeptics.com/forums/showthread.php?t=68227

I really can't see any way to test this highly subjective exchange. With the sex of an unborn baby, there are two possible outcomes (boy or girl); thus, in theory, the psychic could get really lucky and gain the holy grail of 16 out of 20. Having said that, maybe a better test would be to ask the psychic to identify the names of random participants and their parents. Take 40 anonymous participants, and use 20 out of these 40 in the experiment. This way there is much less likelihood of luck saving the day!

 

[Anacoluthon64]

A tarot reader woorth his salt should at least get 16 out of 20 correct to start convincing us, shouldn't he? (See Carina Landin protocol.)

As I understand it, the preliminary test should have a success-by-pure-chance probability of about 1:1,000. If so, the 16 or more correct out of 20 criterion is a bit lenient, having a success-by-pure-chance probability of about 1:169. Also, pregnancies are considerably more common than dead people's diaries (different priorities, you see), so obtaining an adequate number of test subjects shouldn't present any major difficulties. A success criterion of 25 or more correct out of 32 is both manageable and convincing, having a success-by-pure-chance probability of about 1:951.

'Luthon64

 

[nonbeliever]

Yes, a simple pregnancy test sounds good.

Even simpler: If the tarot reader were amenable, thirty or forty people holding envelopes with cards that say "0" or "1" might be enough. The reader could predict which envelopes have which numbers.

Or, since it's easier to find thirty women who already have kids, the tarot reader could determine whether the oldest child of each mother is male or female.

I appreciate your responses; thanks!

 

[drkitten]

Yes, a simple pregnancy test sounds good.

Even simpler: If the tarot reader were amenable, thirty or forty people holding envelopes with cards that say "0" or "1" might be enough. The reader could predict which envelopes have which numbers.

Or, since it's easier to find thirty women who already have kids, the tarot reader could determine whether the oldest child of each mother is male or female.

I appreciate your responses; thanks!

The key, of course, is in finding a tarot reader who will be willing to commit to the ability to perform any of these specific divinations (and to providing sufficiently clear-cut answers). One reason I suggested divining the sex of an unborn child is because that's a "traditional" thing that tarot readers (as well as other forms of soothsayers) have been asked to do. It's a tradition that dates back centuries for tarot and millenia for divination in general.

I don't think there's any "tradition" of divining the contents of a sealed envelope outside of stage magic; a tarot reader could "justifiably" back out of that as being inappropriate or something.

As Randi put it, "I won't ask a violinist to play the piano."

You have to make sure that your proposed test is within the abilities that the claimaint actually claims.

 

[nonbeliever]

Good points; I'll try to keep this in mind. The ideal test should be one that can be proposed by the claimant, agreed upon by the tester, and conducted by a third party.

 

[CynicalSkeptic]

Or, since it's easier to find thirty women who already have kids, the tarot reader could determine whether the oldest child of each mother is male or female.

Nope, not double-blind. A good cold-reader might be able to get info from the mother on the gender of her child.

 

[TheBoyPaj]

I agree with Thomps1d. You sit the reader across the table from the sitter. A screen is placed near to the door of the room, with a table behind the screen. The sitter shuffles the cards and places them on the table in front of him. A third party enters, takes the deck behind the screen and opens an envelope. The envelope either tells him to switch the deck with an identical, preshuffled deck, or just to handle the original deck to make it appear disturbed. They would also have to check the deck for evidence of marking, in case the sitter was cooperating with the reader. The third party leaves the deck on the table behind the screen and exits the room without communicating.

The reader collects the deck and performs the reading. Afterwards, the sitter has one month to evaluate the accuracy of the reading on a scale of 1-10. You would expect true readings to be significantly higher on the scale than the false readings.

The problem I see with this is that there are ways of marking a deck which would be very difficult to spot if you did not know where to look. A fourth party, watching the shuffling taking place, could spot attempts to indent cards with fingernails etc.

 

[ChristineR]

I suppose it depends on the reader, but the usual procedure is to shuffle the cards in the presence of the subject then to let the subject cut the cards.

Afterwards, the subject can leave the room, and the reading can proceed. The tarot reader writes down the results. Repeat this with 10 subjects, then ask the subjects to pick their own reading out of the 10.

You would have to pick 10 similar subjects, e.g., all female, all between 25-30 years old, all white, all raised in the U.S., all in general good health. This keeps the reader from saying things like "all your life you've felt like a second class citizen" based on visual cues.

 

[ChristineR]

I agree with Thomps1d. You sit the reader across the table from the sitter. A screen is placed near to the door of the room, with a table behind the screen. The sitter shuffles the cards and places them on the table in front of him. A third party enters, takes the deck behind the screen and opens an envelope. The envelope either tells him to switch the deck with an identical, preshuffled deck, or just to handle the original deck to make it appear disturbed. They would also have to check the deck for evidence of marking, in case the sitter was cooperating with the reader. The third party leaves the deck on the table behind the screen and exits the room without communicating.

The reader collects the deck and performs the reading. Afterwards, the sitter has one month to evaluate the accuracy of the reading on a scale of 1-10. You would expect true readings to be significantly higher on the scale than the false readings.

The problem I see with this is that there are ways of marking a deck which would be very difficult to spot if you did not know where to look. A fourth party, watching the shuffling taking place, could spot attempts to indent cards with fingernails etc.

There's an easy fix for this one. Use two decks, obviously different from one another. Both are shuffled by the testers, using a machine. Everyone leaves the room. The sitter and a randomizer enter the room. They pick one deck, the sitter shuffles and cuts, they leave. The reader and a second randomizer re-enter the room. They pick a deck, and the reading proceeds. The reader never sees the sitter.

Whether or not this would be acceptable depends on the particular reader and how he/she thinks the whole thing works.

 

[PennyBright]

I think that tarot reading would be very difficult to formulate a protocol for, for precisely the reasons outlined by chillzero and Simon. The infant gender test or child gender test would definitely be good places to start.

I'll follow this thread with interest - I read tarot as a psychological tool, and while I (and from all reports the people I've read for) find it useful and informative, I do not think it is in any way paranormal. It's just a method of re-interpreting past and current events, and re-framing thinking about future ones.