After my fantastic and entertaining chat with Gordon on his Rune Soup podcast, and in tandem with the good Dr Al’s course on the fundamentals and history of the art, there’s been a huge influx of interest in geomancy, to which I say “about goddamned time”. As my readers (both long-term and newly-come) know, I’m somewhat of a proponent of geomancy, and I enjoy writing about it; it’s flattering and humbling that my blog is referred to as a “treasure trove” of information on the art, and I consistently see that my posts and pages on geomancy are increasingly popular. It’s also encouraging enough to get me to work more on my book, which…if I actually get off my ass and work on it like I need to and should have been doing for some time now, will probably get put to consumable paper sometime late next year.

One of the most common questions I find people asking when they first get introduced to the art of geomancy is “how do people generate the geomantic figures?” Unlike other forms of divination, geomancy isn’t tied down to one specific means or method. Tarot and all forms of cartomancy use cards, astrology uses the planets and stars, scrying uses some sort of medium to, well, scry; we often classify methods of divination based on the set of tools it uses, and give it an appropriately-constructed Greek term ending in -mancy. Geomancy is different, though; truly any number of methods can be used to produce a geomantic figure, because geomancy is more about the algorithms and techniques used in interpretation rather than the tools it uses to produce a reading. Once you get into the feel and understanding of geomancy, you can almost quite literally pull a chart out of thin air using any tools (or none at all!) at your disposal. Still, partially because of the ability to be so free-wheeling, newcomers to geomancy are often caught up in the tool-centric way of thinking of divination, and can become (I find) overly concerned with the “best” or “most popular” method.

To that end, let me list some of the ways it’s possible to come up with a geomantic figure. I don’t intend for this to be an exhaustive list, but more of a generalized classification of different kinds of ways you can produce a geomantic figure (or more than one in a single go):

**Stick and surface**. This is the oldest method, going back to the very origins of the art in the Sahara, where the geomancer takes some stylus and applies it to an inscribable medium. You can use a staff and a patch of soil on the ground, a wand on a box of sand, a stylus on a wax (or modern electronic) tablet, a pen on paper, or some other similar mechanic. To use this method, simply make four lines of dots, traditionally from right to left. Don’t count the dots; let them fall naturally, so that a random number of dots are in each line. Some people get into a trance state, chant a quick prayer, or simply focus on the query while they make the dots, if only to distract the mind enough to avoid counting the dots and influencing what comes out. Once you have four lines, count the dots in each line; traditionally, the geomancer would cross off the dots two-by-two (again, right-to-left) until either one or two dots were left over at the end. These final leftover dots are then “separated” out from the line to form a single figure. To make all four figures, simply increase the number of lines from four to sixteen, and group the rows of leftover dots into consecutive, non-overlapping groups of four rows.**Coins**. This is a simple, minimalist method: flip a coin four times. Heads means one point of the resulting figure, and tails means two (or you can swap these around, if you so prefer, but I prefer heads = one point). Flipping a coin four times gets you four rows to make a complete figure. Alternatively, you could flip four coins at once, perhaps of different denominations: for example, you could flip a penny for the Fire line, a nickle for the Air line, a dime for the Water line, and a quarter for the Earth line; a single throw of all four coins at once gets you a complete geomantic figure. I consider any method that uses a “flip” to produce a binary answer to fall under this method; thus, the druid sticks used by geomancers like John Michael Greer and Dr Al Cummins would technically be considered a type of geomancy-specific “coin”, as would pieces of coconut shell where the convex side on top is “up” and the concave side on top is “down”.**Divining chain**. This is a slightly modified version of the coin-based method, where four coins or disks are linked together in a chain. Rather than throwing the coins individually, the chain itself is flung, tossed, or thrown in such a way that each coin falls on a different side. The only example I can find of this in Western-style divination is the (possibly spurious) Chain of Saint Michael, where four saint medallions are chained, one to another, and connected to a sword charm, but a corollary to this can be found in the Yoruba divination methods of Ifá, using something called the ekuele (or ekpele, or epwele, depending on whether you’re Cuban or Nigerian and how you feel like spelling it). There, you have four pieces of cut shell that can fall mouth-up or husk-up, or four pieces of metal that fall on one of two sides; notably, the ekuele has eight coins on it so that the diviner-priest can throw two figures at a time, but that’s because of the specific method of Ifá divination, which is only a distant cousin to geomancy and*shouldn’t actually be mixed with our techniques*.**Dice**. Again, a pretty straightforward method: roll a single die four times, or four different dice one time. If a given die is an odd number, use a single point; if an even number, use two points. Some people use four different-colored cubical dice (e.g. red for Fire, yellow for Air, blue for Water, green for Earth), but I prefer to use tabletop RPG dice that come in different shapes. For this, I use the associations of the Platonic solids to the classical elements: the tetrahedron (d4) for Fire, octahedron (d8) for Air, icosahedron (d20) for Water, and cube (d6) for Earth. Like Poke Runyon aka Fr. Therion, you could use four knucklebones for the same purpose, as each knucklebone has four sides (traditionally counted as having values 1, 3, 4, and 6). Dice are easy, the tools fit in a tiny bag which can itself fit into a pocket, and nobody is any the wiser if you just pull some dice out and start throwing them on a street corner.**Counting tokens**. This is a similar method to using dice, but a more general application of it. Consider a bag of pebbles, beans, or other small mostly-similar objects. Pull out a random handful, and count how many you end up with. If the number is odd, give the corresponding row in the geomantic figure a single point; if even, two points. This is a pretty wide and varied set of methods; you could even, as Nigel Pennick proposes, pull up four potatoes from a field and count whether each potato has an odd or even number of eyes on it. The idea here is to use something to, again, get you a random number that you can reduce into an odd or even answer, and isn’t really different from using dice, except instead of being presented with a number, you have to count a selection of objects obtained from a collection. In a sense, both the dice and counting token methods can be generalized as using any random-number generator; you could use something like random.org to get you four (or sixteen) random numbers, to which you simply apply the odd-even reduction; such a generator can be found using this link.**Quartered drawing.**Not really a technique or toolset on its own, but a variation on things that use coins, identical dice, or other counting tokens. In this, you prepare a surface that’s cut into four quarters, such as a square with four quadrants or a quartered circle. Each quarter is given to one of the four elements, and thus, to one of the four rows of a geomantic figure. Into each quarter, you’d randomly flip one of four coins or drop a random number of beans, and read the pattern that’s produced as a single figure. This can be useful if you’re short on similar-but-not-identical tools (like only having four pennies instead of four different types of coin, or four identical dice instead of different-colored/shaped dice).**Selection of numbers**. One method of geomantic generation I know is used in Arabic-style geomancy is to ask the querent for a number from 1 to 16 (or, alternatively, 0 to 15). Arabic-style geomancy places a huge emphasis on taskīn, or specific orders of the figures which are correlated with different attributions; one such taskīn, the Daira-e-Abdah, simply arranges the geomantic figures numerically, using their representation as binary numbers. From the Ilm-e-Ramal group on Facebook, here’s a presentation of this taskīn with each figure given a number from 1 through 16:

Personally, I use a different binary order for the figures (reading the Earth line as having binary value 1, Water as binary value 2, Air as binary value 4, and Fire as binary value 8), where Populus = 0 (or 16), Tristitia = 1, Albus = 2, and so forth, but the idea is the same. To use this method, simply get four random numbers from 1 to 16 or (0 to 15), and find the corresponding figure in the binary order of the figures. You could ask for larger numbers, of course; if a number is greater than 16 (or 15), divide the number by 16 and take the remainder. You could use dice to produce these numbers, or just ask the querent (hopefully ignorant of the binary order used!) for a number. In fact, you’re not bound by binary ordering of the figures; any ordering you like (planetary, elemental, zodiacal, etc.) can be used, so long as you keep it consistent and can associate the figures with a number from 1 to 16 (or 0 to 15).**Playing cards.**A standard deck of 52 playing cards can be used for geomantic divination, too, and can give that sort of “gypsy aesthetic” some people like. More than just playing 52-Pickup and seeing whether any four given cards fall face-up or face-down to treat cards as coins, you can draw four cards and look at different qualities of the cards to get a different figure. For instance, are the cards red or black, odd or even, pip or face? With four cards, you can make a single figure; with 16, you can make four Mothers. Better than that, you can use all the different qualities of any given card of a deck to generate a single figure, making the process much more efficient; I’ve written about that recently at this post, which you should totally read if you’re interested. What’s nice about this method is that you can also use Tarot cards for the same purpose, and some innovators might come up with geomancy-specific spreads of Tarot that can combine the meanings of the Tarot cards that fall with the geomantic figures they simultaneously form, producing a hybrid system that could theoretically be super involved and detailed.**Geomantic tokens**. Some geomancers have tools that directly incorporate the figures, so instead of constructing a figure a line at a time like with coins or beans, a whole figure is just produced on its own. Consider a collection of 16 tokens, like a bag of 16 semiprecious stones (like what the Astrogem Geomancy people use), or a set of 16 wooden discs, where each token has a distinct figure inscribed on each. Reach into the bag, pull out a figure; easy as that. If you use a bag of 16 tokens and are drawing multiple figures at once, like four Mothers, you’ll need to draw*with replacement*, where you put the drawn token back into the bag and give it a good shake before drawing the next. Alternatively, if you wanted to draw*without replacement*, you’ll need a collection of 64 tokens where each figure is given four tokens each, such as a deck of cards where a single figure is printed onto four cards.

As for me? When I was first starting out, I used the pen-and-paper method (or stick-and-surface method, to be more general). This was mostly to do a sort of “kinetic meditation” to get me into the mode and feel of geomancy, going back to its origins as close as I could without being a Bedouin wise-man in the wastes of the Sahara. After that, I made a 64-card deck of geomancy cards, with each figure having four cards. I’d shuffle the deck, cut it into fourths from right to left, and flip the top card of each stack to form the Mothers. For doing readings for other people in person, like at a bookstore or psychic faire, I’ll still use this; even if geomancy isn’t familiar to people, “reading cards” is, so it helps them feel more comfortable giving them a medium they’re already familiar with. Plus, I also can get the querent’s active involvement in the divination process by having them be the ones to cut the deck after I’ve shuffled; I’ll still flip the top card, but I find having them cut the deck gives them a meaningful inclusion into the process. Generally, though, I use tabletop RPG dice for the Platonic solids. I roll the dice and see whether each die is odd or even for a single figure, so four throws of dice get me four Mothers. Nowadays, I only use the stick-and-surface method if I have truly nothing else at hand, because I find the process to be slow and messy, but it still works, and I can still rely on my own familiarity with it so that it doesn’t trip me up when I have to use it.

What would I suggest for newcomers to the art? Like me, I’d recommend new geomancers to start with the stick-and-surface method, if only to develop an intimacy with the underlying, traditional method that produced all the others. In a sense, doing this first is like a kind of initiation, practicing the same fundamental technique as have geomancers for a thousand years, and itself can be a powerful portal into the currents of the art. Once you have that down-pat and have gotten into the feel of the art, though, I find that the method is pretty much up to the desires and whims of the geomancer. Anything that returns a binary answer can be used for geomancy, but for convenience, some people might prefer instead a “whole figure” type of draw. Once you settle on a set of tools, for those who are of a more magical or ritual bent, you may want to consider consecrating or blessing them, or entrusting them to the connection and care of a divining or talking spirit, according to whatever methods you find appropriate, but this isn’t strictly necessary for the art, either.

Ultimately, the tools you use for geomancy are entirely up to you, because it’s the techniques and algorithms we use that are what truly makes the art of geomancy. The only thing I really recommend is that the geomancer takes an active role in divinely manipulating the tools used to produce the figures.

How about you, dear reader? What methods do you use for geomantic generation? Have you heard of any that aren’t on the list above, or aren’t included in any of the above classifications? What are you most comfortable with? What methods do you dislike, either on a practical or theoretical level? What would you recommend?

I started with coins. I’d worked a few years back on the I Ching and while I never felt like I memorized the key elements of the figures enough to really get it to work for me, the coin method for generating hexagrams stuck around. I then started working with an RPG dice roller I have on my phone, just because it was always available. But wanting to get more hands on again, and inexplicably having plenty of DnD dice lying around, I started using your platonic solids method. I like loading the elemental symbolism in from the jump.

I’m also attracted to the idea of random things: leaf piles, how many cars go by an intersection in a set amount of time, &c. But that often requires that I ask questions on a walk, which is an amazing ideal thing that never quite happens.

I’m going to give the dots on a page method a go over the next few days, just to see. It always felt like I’d never get to a random number that way, that I’d always be too involved in the process, but we’ll see: I don’t have access to potato fields on the regular.

Awesome stuff, man. Thanks!

One of the reasons why I like recommending the stick-and-surface method for beginners is because of that very fear: not only do you have to learn to trust the method, you also have to force yourself not to count, and get more into the mindset/headspace of a proper geomantic “trance”, as it were. Getting into that trance takes practice, but once you get there, it serves you in every other method. The other methods don’t force that upon you in the same way from the get-go.

I’ll mention this on the rune soup forum as well, but I took your advice and worked with the stick-and-surface method: exactly as advertised. Since getting into some kind of trance state is familiar, this was remarkably easy and refined my headspace very quickly. I got excellent and unpredicted results. Cheers!

I like to take ask my question and then allow my senses to gently scan my surroundings and quantify whatever catches my attention . These numerical results are then “translated” into binary results.

If I see a feather it could be “uneven” (1 feather), but if I notice the two sides of barbs to the right and left of the rachis instead, then it would be “even” (2 sides). A Flock of brids could be uneven (1 flock) but if the individual birds catch my attention I count them and might end up with an even number of birds. Sometimes I notice a word being repeated x nuber of times in whoever talks to me first after I ask my question. Sometimes have to sneeze a number of times. It all boils down to let the anima mundi speak to you through your senses and then translate her song into geomantic figures.

I am still looking for someone who can teach me on how to know the outcome of a football match with geomancy