There is no official Bananagrams dictionary

Many players like to cite the Official Scrabble Players Dictionary to validate words they use when playing Bananagrams. But as discussed in the last post, the OSPD has serious flaws and may not be the best choice for a reference. The official Bananagrams rules say that "any available dictionary may be used" to decide whether words are acceptable. Some better dictionaries to use with Bananagrams are listed here.

As an alternative, when playing Bananagrams among friends, the group can decide how strict it is going to be about word acceptability. A strict reading of the rules would say that if a word is in the dictionary, it is acceptable, whether or not it is slang. In the groups that I play, we are pretty lax, and frequently we do not even have a dictionary on hand. When playing without a readily accessible dictionary, the validity of words is sometimes debated, but the de facto standard is one of the following (I'm not sure which): 1) The word is only considered a rotten banana if it is misspelt or clearly not a word. 2) The word is considered a rotten banana if most people think it is wrong. 3) The word is considered a rotten banana if the person who played it relents and agrees that it is wrong (or questionable).

---

Acknowledgments: My thanks to Chuck, who started an e-mail discussion with me which led to this post and a deeper consideration of dictionaries.

Well, that about wraps it up for the Official Scrabble Players Dictionary

I found a nice critique of the Official Scrabble Players Dictionary in one of the reviews on Amazon. It's by Daniel Pratt, a lexicographer and mathematician who placed second in the first national Scrabble tournament in 1978. He also developed the rating system used for Scrabble players in tournaments. It was around the same time that the first Official Scrabble Players Dictionary was created. Pratt is therefore well qualified for making such a critique.

One of his strongest points is something that I had begun to suspect: The OSPD adds new words when they appear in dictionaries, but does not delete old words once they become obscure enough to not appear in modern dictionaries:

The Official Scrabble Players Dictionary (OSPD) is a compilation of words from twelve U.S. college dictionaries from the last four decades. Four are still in print and, as the descendants of seven of the others, contain most but by no means all of their contributions. As a result, pronunciations, etymologies, and full definitions are no longer available for many entries, especially those found only in the source that has been out of print for a quarter century. NSA members like to twitter that they don't play your grandmother's Scrabble, but in many respects they're using her dictionary. It's a shame they can't bring the game into the 21st century.

He also points out some of the most ridiculous entries, but the word that sticks out most in my mind is one that was pointed out elsewhere: "AARRGHH" is listed in the OSPD, like it is an actual word.


Ammon Shea takes a different position. After reading the entire Oxford English Dictionary (which feat he documented in his book, Reading the OED: One Man, One Year, 21,730 Pages), he describes returning to Scrabble for the first time in many years and how he finds that the list of allowed words no longer makes sense to him:

Why, for instance, does the Scrabble Players Dictionary list both howf and howff (Scottish words defined as ‘a place frequently visited’), but neglects to include many of the variants that Joseph Wright has in his English Dialect Dictionary, such as houf, hauf, hofe, hoff, houf, houck? I tried to play swad (a bumpkin, or fat person), which appears in most of the dictionaries I own, on three different occasions before I remembered that the Scrabble game wouldn’t recognize it.

He then goes on to observe:

Some complain that the Scrabble Players Dictionary is too inclusive; I find the opposite to be true. Rather than only let in some of the strange words, they should have opened the floodgates and allowed them all. It has become a game of memory, rather than a game of language.

Clearly, you can't please everybody.

---

In response to comments on his review, Pratt recommends several replacement dictionaries - the American Heritage College Dictionary, the American Heritage High School Dictionary, the New Oxford American Dictionary, and the Canadian Oxford Dictionary - with a discussion of the advantages of each.

An article about the makers of Bananagrams

The Boston Globe has published a nice article about the invention of Bananagrams, giving more details than I have ever previous encountered, a photograph of the Nathansons (the family that invented the game four years ago), and quotes from Abe Nathanson himself.

The article describes how different family members contributed to different aspects of Bananagrams and how many of the details of gameplay were developed by Abe while working on his own over the course of almost a year.

Abe's aversion to the big retailers and the way they do business is the reason that Bananagrams is not sold at stores such as Toys R Us or Wal-Mart. Similarly, the Nathansons have no interest in selling the rights to Bananagrams to any of the big toy companies. Their success in independently developing a game from concept to best-selling juggernaut makes for an interesting story.

The last paragraph of the article is a hint about future directions of the company:

"We're almost at the point where we have more money than we need," says Nathanson, adding that he's drawing the line at creating any more fruit games. Bananagrams will always be the gold standard, he says, "and we don't want to kill a good idea."

So, no more new fruit-based games. Next up: Cabbagegrams! Lettuce Play! Beet, Don't Fail Me Now!

Bananagrams on Twitter

October 2011 Update (superseding the original post which is below): The official Bananagrams Twitter account has changed from _BANANAGRAMS_ to bananagrams.

---

Large Animal Games, the makers of the iPhone and Facebook version of Bananagrams, has been been posting updates on Twitter as bananagrams since early 2009. (The first post is here.)

I occasionally post Bananagrams-related things under the Twitter alias bananagrammer.

And now, there is an official Twitter feed for Bananagrams:

_BANANAGRAMS_

_BANANAGRAMS_ has started posting "Weords", a term introduced in the
Bananagrams book, meaning "weird words that are cool to play". It seems like it will be a useful account to follow. I am looking forward to future posts.

On the effect of body orientation on anagramming speed

When we are standing up, gravity pulls blood down which is detected by "baroreceptors" in the circulatory system. The circulatory system compensates by increasing the heard rate, so that blood pressure in the upper body is maintained. When the baroreceptors register such a blood pressure drop, the locus coeruleus neurons become more active. The locus coeruleus responds to stress by increasing release of norepinephrine (a.k.a., noradrenaline). This has a number of effects, including increasing heart rate and blood flow and boosting motivation, thinking, and alertness. But researchers have found that increases in norepinephrine coincide with a slight slowing of certain kinds of thinking.

The thesis is that "insight problems", the kind where your brain kind of wanders and does background processing, and then suddenly presents a solution in an "A-ha!" moment, happen faster when you have lower norepinephrine levels. A simple "A-ha"-type problem to test, and one which is very popular in psychological research, is anagramming.

In one experiment [M.P. Walker, C. Liston, J.A. Hobson, R. Stickgold, "Cognitive flexibility across the sleep-wake cycle: REM-sleep enhancement of anagram problem solving", Cognitive Brain Research, Volume 14, p.317-324 (2002)], researchers woke people up at different stages of sleep and asked them to solve anagrams. They found that subjects who had just been in REM sleep were better at anagramming than subjects who had been in some other sleep stage. And REM sleep has a lower locus coeruleus activity level.

In another experiment, [D. M. Lipnicki and D. G. Byrne, Cognitive Brain Research, Volume 24, p.719-722 (2005)], researchers had subjects solve 5-letter anagrams (like unscrambling DEFSU to get FUSED) and perform arithmetic problems (73 - 58 + 19 - 26 = ?), while standing up and then while lying down. When standing up, average anagramming time was 29.4s +/- 6.3s. When lying down, average anagramming time was 26.3s +/- 5.4s. The difference in anagramming times (3.1 seconds) is smaller than the standard deviation of the data, but the statistical analysis suggests that the error bars are not that important, allowing us to conclude that the average person in the study anagrammed 10% faster while lying down.

A similar test found that standing is correlated with people solving simple arithmetic problems a few percent faster, but it's such a small difference that it doesn't seem statistically significant.

Summarizing: When you stand up, your body tends to become more alert and less relaxed, and you may anagram slightly slower as a consequence.

Caveats: This study was performed on randomly selected subjects with no proclivity for word games. When given 32 5-letter anagrams to solve, and 45 seconds to solve each in, the average number of solutions was 8 or 9 (whether sitting or standing). A typical Bananagrams player would do significantly better, I would be willing to bet.

I wonder if the cause and effect might flow the other way: if lower blood pressure allows one to anagram faster, can focusing on anagramming cause blood pressure to be temporarily lowered? Or will it just cause one to lie down?

Anagramming is only one of the skills used in playing word games, of course. The kind of concentration necessary to play Bananagrams fast most likely requires the mind to not be in a relaxed state. (And what I'd really like to see is a study on playing Bananagrams itself). Still, whenever the professional Bananagrams league launches, one of the first drugs they should consider regulating is beta-andrenoceptor antagonist propranolol.

---

Further reading:

D. M. Lipnicki and D. G. Byrne, "Thinking on your back: Solving anagrams faster when supine than when standing", Cognitive Brain Research, Volume 24, p.719-722 (2005).

The full paper can actually (to my complete surprise) be bought through Amazon, or if you have access to it through your library, you can get the paper from here.

How to remember the number of tiles to start off with in Bananagrams

The simplicity of Bananagrams is one of its strong points. But there is one rule that always seems to cause debate and searching for the instructions: The number of tiles that each player is supposed to begin with. The rules call for the following initial tile counts:

• 2-4 players - 21 tiles
  
• 5-6 players - 15 tiles
  
• more than 6 players - 11 tiles

I find it easy to remember the cutoffs for each level, probably because 4 and 6 are both even. And 21, the number of tiles when playing with 4 players, is burned into my memory. So here is my mnemonic for determining the other numbers: After the cutoff for 4 players, subtract 6 tiles. After the cutoff for 6 players, subtract 4 more tiles.

That should shave ten seconds off my next Bananagrams match. Plus now I am no longer bothered about losing my set of Bananagrams instructions.

Tactics to help you win at Bananagrams

Through playing many, many games of Bananagrams, I have learned a few lessons about how to approach the game:

• Concentrate on making words out of the hardest letters first to get them out of the way.
  
  
• Try to make one long word at the beginning and build off of it. In general, the longer your words are, the more places there will be to build new words.
  
  
• Particularly at the beginning, when you flip over all your letters, try to assess the balance between consonants and vowels. If it is off, you can try to make vowel-heavy or consonant-heavy words [See what to do when you have too many consonants and what to do when you have too many vowels.]. It's far better to do this from the outset, than to have a mostly-finished grid and realize that you need to go back and remodel the whole thing.
  
  
• Try to structure your grid in a more open way.
  
  Good:

     CHILL      
         I 
         POST

  Not-as-good:

     CHILL
         I
      STOP

  The former is far easier to build new words off of:
  

     A
     CHILL W
     U   I A  
     T   POST
     E      O

  We could even quantify the difference between the two approaches. In the first case, words can be build up or down (or both) from 5 letters (C,H,I,S,T) and just in one direction from 2 more (the L and the O). In the second case, there are 8 letters that one can build off of in only one direction. The number of words you can make with the H in any position (HUE, OCHRE, EACH) versus just making words that start or end with a given letter, is definitely bigger than a factor of 2... I'm going to conservatively estimate it to be 3 times larger. Then the flexibility scores of the two grids above would be 3*5+2=17 and 8*1=8. And the advantage of choosing an open structure only grows as the grid becomes larger. It's often the difference between smoothly adding letters to your grid and having to completely rearrange things.
  
  
• When you are in the peeling phase, keep in mind one or two words that you can steal letters from and positions where you can easily add tiles to make new words. Or words that you can easily anagram (with the newly peeled letter) to make a new word.
  Suppose you have the grid below and you peel an A.

       TRY
       I
       E
     FORKS

  From FORKS, you can take the S and combine it with the A to from SAY:

         S
         A
       TRY
       I
       E
     FORK

  Now that I'm thinking about FORK and SAY, I note that the K from FORK could be substituted for the more-easily-used A in SAY to form SKY. So if the next letter I peeled was R, I could quickly make that switch, freeing up the A to build the word FAR:

         S
         K
       TRY
       I
       E
     FOR
     A
     R

  At this stage, I might wind up extending FOR to FORT or FORD or FOUR or FORE or anagramming it to ROOF, depending on what letter become available. It would be nice to build SKY into RISKY. And FAR looks like an appealing place for making FARM, FAIR, or FARE.
  

The speed of Bananagrams is such that you will not be able to consider every possible move, but having just a few rules of thumb to keep you from getting boxed in can allow you to still play fast while increasing your chances of winning.

Another reason Scrabble dictionaries are inadequate for Bananagrams

The Scrabble board is a 15-by-15 grid, so naturally, Scrabble dictionaries have no need for words longer than 15 letters. Oddly even "Super Scrabble" which is Scrabble played on a larger board and with more tiles, still uses the 15-letter limit.

Some 16-letter words for your consideration:

ambidextrousness
anagrammatically
anthropomorphize
autobiographical
biodegradability
bureaucratically
counterclockwise
crystallographer
cryptozoological

I can totally see someone staring off with "clocks" in their grid, turning it into "clockwise", and then in a brilliant burst of insight, combining it with "count" and some stray tiles to make "counterclockwise". If I ever pulled that off, I would be so proud that I wouldn't even care if I won the game. (OK, I'd still care, but either way it would be awesome.)

What is the longest word that you have ever used in a Bananagrams match? For those who like games with rounds that escalate, the must-contain-an-N-letter-word would be an interesting and challenging variation. How high can you go?

Scrabble Me: The chimera of Scrabble games

I recently ran across the strangest official Scrabble mutation that I have ever seen: "Scrabble Me"

It looks like, in addition to Scrabble Apple, the makers of Scrabble decided to make another game that fuses elements of Bananagrams with those of conventional Scrabble. This one is kind of a Frankenstein hybrid, so don't say I didn't warn you:



The game is divided into rounds. In each round, everyone uses up their tiles by building words on independent mini-Scrabble-boards. Then players score their words, draw new tiles, and continue building up their grids.

Unlike Scrabble Apple, which seemed to be fairly close to Bananagrams, with some Scrabble elements thrown in (scoring and double word score squares), Scrabble Me seems like it is deviating from the Scrabble - Bananagrams continuum. It's more random, in the sense that when someone uses a blank tile (called "wild" in this game), they have to swap their grid with someone else's. However, it is less random in the sense that there is a pool of face-up tiles that one can opt to draw from, instead of from the bag. It somehow reminds me of the card game Uno.

In the interest of trying to categorize these games, I've made this ASCII chart:

   ^                     BANANAGRAMS
S  |
P  |
E  |
E  |
D  | SCRABBLE          SCRABBLE ME
   |
   |---------------------------------->
         DECOUPLING OF PLAYERS        

And there really ought to be a third-dimension where Scrabble Me towers above the others on the Uno-ness axis.

The reviews are sort of favorable though. In spite of the Frankenstein quality, I would like to try this game.

Palindromes you are most likely to be able to make in Appletters

Since making palindromes earns players bonus points in the Applescore game (as described in this previous post on Appletters), I've compiled a list of the palindromes you are most likely to be able to make during the game (really short ones or ones that use common letters).

Off the top of my head, using only the minimal length examples:

bib, bob
dad, deed, did, dud
eke, ere, eve, ewe, eye
gag, gig
kook
mom
noon, nun
pap, pep, pip, pop, pup
radar
sees
tat, tit, tot
wow

And, of course, there is one Scrabble-legal letter combination that is two letters long and a palindrome: AA. A'a (pronounced /ah ah/) is a type of lava, which is thicker and more viscous than other types. It is characterized by flowing in a sporadic fashion and leaving a rough surface when it cools.

And if it's not already been suggested, then I propose a variation of Appletters called Palindrominoes: The word snake is formed as in regular Appleletters, but if you can form a palindrome where the base letter (the "head" of the snake, I suppose) is inside the palindrome, then you are permitted to position letters both above and below the end-tile, as shown in the example below:

  e   
  VIGOR 
  e   I
      N  
      KITE

Then you can give the snake multiple heads, like a hydra or a planarian or something. A little extra chaos to spice up your Appletters game.

The games of Lewis Carroll

In addition to his anagramming prowess and his wonderfully surreal and whimsical stories, Lewis Carroll is known for creating numerous games and puzzles.

He is credited with the invention and popularization of the doublet (a.k.a, Word Ladder) in which one transforms one word into another, one letter at a time, with all the intermediary steps being legitimate words. Two examples:

          EAST          CAT
          FAST          COT
          FEST          COG
          WEST          DOG

Admittedly, this is really more of a puzzle or a pastime than a game.

He spent much of his professional life tutoring students in mathematics and logic. He invented a game called The Game of Logic designed to teach the fundamentals of formal logic, using a unique way of representing logic propositions with a game board and colored tokens. The game is described in a book of the same name (available from Project Gutenberg (though they seem not to have gotten the figures right) and Amazon). Ultimately though, it seems to be for one player and seems rather like a puzzle.

Carroll also thought about and devised rules for playing billiards (the British sort, played on a table without pockets) on a circular table.

In 1880, he wrote in his diary that "A game might be made of letters, to be moved about on a chess-board till they form words.". Martin Gardner (famous for his writings on recreational mathematics) took such notes from the writings of Carroll and fleshed them out into a board game in which, as I understand it, letters are placed on the first row of a chessboard and can be moved like queens to form a word in the fifth row. Meanwhile your opponent is trying to do the same thing while blocking you. (Rows 2, 3, 6, and 7 are open to everyone.) Like Scrabble, once a word has been formed, more letters are drawn from a bag, and the first row is replenished. You can buy the game here. I am thinking of playing an improvised version with a chessboard and a set of Bananagrams tiles.

History of anagrams: Anagramming as recreation and social skill

Anagrams were very popular during the 17th century, and particularly in France. King Louis XIII used anagrams as a recreational form of intellectual stimulation and source of entertainment for himself and members of his court. He formed some on his own and hired a Royal Anagrammatist whose sole task was to devise anagrams. A lot of these anagrams were based on the names of people, and particularly, on the name of the king with the theme being how great he was.

The book "Of Anagrams" by Henry Benjamin Wheatley (downloadable for free from its Google Books entry) says that "Anagrams were used by fine gentlemen to add pungency to their conversation". It cites Henry Peacham's "Compleat Gentleman" (written in 1634) which basically says that among the conversational skills of a gentleman ("merry tales, wittie questions and answers", "ingenious epigrams") should be the spontaneous spouting of anagrams. He gives the example of someone who was going to say "I must goe buy a dagger" but decided to switch around the letters and say "I must goe dye a beggar". [It's not strictly an anagram though; it's actually a rearrangement of sounds. A more precise description might be "anaphone".]

In the 19th century, it was popular to anagram the names of famous people (celebrities, I suppose). Lewis Carroll produced many such anagrams, including Florence Nightingale ("Flit on, cheering angel."), Disraeli ("I lead, sir."), and a politician of the time named William Ewart Gladstone ("Wild agitator means well!"). The popularity of anagramming at the time led to the Victorian game of Anagrams (previously described here and which now is sold under the name "Snatch") which was played by Alfred Butts as a child. His first attempts at inventing a game drew inspiration from the idea of anagramming as the basis of a game, to which he added a letter distribution like that of the English language. After a few more iterations, he converged on the game of Scrabble, which led to Speed Scrabble, which led to Bananagrams.

Review of the Bananagrams iPhone app

The Bananagrams application for iPhone and iPod touch is now available [link to iTunes Store] for $4.99. [Update (03/2010): It is on sale for 99 cents.] It is pretty much an exact port of the Bananagrams game on Facebook and seems to differ only in the interface.

When playing offline, you can play the two one-player games: Solo Cafe (which is timed and involves forming a grid from 21 tiles, with no peeling (though dumping is an option)) and Solitaire (also timed, but requires peeling to finish the 36 or 72 tile bunch).

You can move around to different parts of the board by putting two fingers on the screen and dragging the board. You can also zoom into or away from the board by using the standard iPhone pinching motion.

If you put your finger right on top of a tile to select it, it should pop up a little bit above your finger tip so you can see what letter you are holding. (It leaves behind a white square outline to indicate the actual tile position). You then drag the tile (in pretty much exactly the same way that you use a mouse to drag a tile in the online version of the Bananagrams game) and then remove your finger off the screen to drop the tile where your finger was last touching.

One improvement over the Flash version of Bananagrams is that the iPhone version takes advantage of multi-touch technology by allowing you to select multiple individual tiles at the same time. Put one finger on one tile, a second on another tile, and a third on yet another tile, and you can move them all around independently. I confess that I have not been able to work this into my game yet, but this feature has serious potential.

It also has the ability to select a group of tiles at the same time, which works pretty much like selecting a bunch of file icons in a typical computer operating system graphical interface: 1) Click on the screen somewhere (by pressing your finger). 2) Drag somewhere else to form a rectangle, bounded by the first and last points. 3) Everything inside has been selected and can be (for instance) dragged simultaneously.

I think that there is a bug in the current implementation though, since there is no visual indication that the rectangular selection is taking place; you just have to imagine it. Once you have completed the selection, the selected tiles light up (slightly). Also, the option to rotate the selected tiles (which I described here for the Facebook version) is currently unavailable.

The game works in both landscape and portrait orientations, though I find that I have to start off in the portrait orientation (so there are the initially overturned tiles on the top and my grid on the bottom). After the grid has formed and the peeling has begun, it sometimes feels more natural to rotate the device and switch to the landscape orientation (depending on the shape of the grid).

As with the online version, if you accumulate enough points you can acquire different tile sets and backgrounds (as shown in the example above).

[At this point, I decided to consider the game for a while longer, ultimately returning six months later to finish this review.]

It would be nice if you could keep playing music while playing Bananagrams. I think that if a song is playing when Bananagrams is launched, you should have the option to suppress sound effects and listen to what you want.

When playing offline, the program does not retain your best time. When playing solitaire Bananagrams (which is a pretty good simulation of an actual Bananagrams game if you play the 36-tile version), I like to time myself and try to beat my own best time. Offline statistics-keeping would be a nice feature.

The online functionality works a bit fitfully over a slow Internet connection. The game needs to transfer things like icons (for player avatars) and background images for the board. It can take a while to transmit game results (which is surprising since very little data needs to be sent). Sometimes a dialog box would pop up saying that the game had lost the network connection, and other times I would play all the way through, and then at the end, the game would fail to report the results to the central server. If you have a decent Internet connection, this should not be a problem.

Other bugs: I got into this situation where I could not scroll the board any further to the right, and my letters were right up against this edge. A workaround is to grab all the tiles and move them away from the edge of the board. Also, though I clicked on the "Remember Me" option for logging into my account, the program has forgotten my login information at least once so far. (It is the e-mail address that you need to log in with, not the name that you register under.)

I'm not so crazy about the size of the iPhone screen. You can resize the game so either you can see the whole board simultaneously but the tiles are small (and sometimes hard to select) or so the tiles are big enough to easily select (like, you can see them pop up from around your fingertip) but then you have to pan around to see different parts of the game. The iPad, on the other hand, will probably be the ideal platform for this game once Large Animal Games releases an iPad-optimized version (not that I know that this is under development).

The final verdict: I am probably not in the target market for this game. I love Bananagrams, but I don't play iPhone games very much. Still, I sometimes fire up the app and play a few rounds of the 36-tile version of "Bananagrams Solitaire". It's OK for that. Who this game is really for is Facebook Bananagrams addicts and people with fast Internet connections. If you fit into these two categories, the bugs described above will probably fade from memory as you get caught up in the online Bananagrams slugfests which is where the action really is.

In closing, "Carpe aríenam!" [Seize the banana!].

Description of the Pairs in Pears game

Now that the new Pairs in Pears game is available, some new information has surfaced. The tiles come in four designs: solid, outline, lines and dots. The different patterns function like different suits of cards. As there are four suits, each containing a complete alphabet, there are a total of 104 tiles. Just as in a deck of cards, each tile is unique (e.g., the G of solids or the H of dots). This adds an interesting new dimension to play with.

The "Pairs in Pears" game involves making pairs of intersecting words that consist of tiles of all the same pattern (as shown in the picture). Whoever makes a certain number of word pairs first, wins. "Pairpoints" is a variation that includes a more elaborate scoring scheme.

I will post more information about gameplay here, as I find it.

Appletters details revealed!

The online version of the Toy Directory trade magazine reports that Appletters (one of the new games from the makers of Bananagrams, previously mentioned in this post) will come with 110 tiles in a cloth apple and will come with instructions for three separate games:

In Appleletters, for two to six players ages 5 and up, players alternately add tiles to the first or last letter of a word in the middle of the table, creating a continuous "snake" of new words. Apple Turnover, for two to four players ages 7 and up, is similar to Appleletters. However, each player begins with 21 tiles instead of nine, and may actually replace an opponent's word with a longer word. The goal is to be the first player to get rid of tiles. In Applescore, for two to four players ages 7 and up, players build words as long as possible in crossword-like fashion and get bonus points for length, palindromes and going out first.

If the spelling above is correct, this means that the overall physical package is called "Appletters", while one of the three games that can be played with the equipment is called "Appleletters". We'll see... All of the games sound interesting. Applescore, in particular, sounds like a sort of Bananagrams-Scrabble hybrid that incorporates the emphasis on long words and scoring of Scrabble while retaining the free-flowing nature, individual grids, and speed of Bananagrams. Adding extra points for palindromes is probably the twist I like the most.


UPDATE: According to the rules of Appleletters, players build a single zigzagging chain of words, like the formation used in dominos:

          TOCK
   GIANT  I  E
   O   A  C  Y
   A   BOOK
   L

The object of the game is to be the first to use all your letters. If, on a given turn, you cannot add a word to one of the two ends, you must draw three more tiles.

The tiles are about twice as thick as Bananagrams tiles, so they can stand up on their own without need for a rack.

Appletters is now available.

Word Game Poetry

The idea of rearranging a set of Scrabble tiles to form a poem has been around for a long time. James Ernest of Cheapass Games is credited with popularizing Scrabble poems.

Here is an example of his Scrabble poetry, formed from 98 letters and 2 blanks:

FANATICS
YOU SEE
HAVE TAKEN OUR QUIET BURG

THEY CLIP POWER LINES
JAM _ADIO AND TV

FIRE BLAZING NEXT DOOR

WOE

I MIS_ GERALDO

This is obviously a work of fiction since no one really misses Geraldo.

I found a six stanza poem, in iambic pentameter, with each stanza made up of the 100 tiles of a Scrabble set. The first one reads:

THROUGH SENTIENT, GAUZY FLAME I VIEW LIFE'S DREAD,
QUIXOTIC, PARTIAL JOKE. WE'RE VAPOUR-BORN,
BY LOGIC AND EMOTION SEEN AS deAD.

It is an amazing achievement in Scrabble poetry.

Inspired, I have written what I believe to be the first Bananagrams poem, using all of the 144 Bananagrams letters.

I call it "Box in Woks".

I SMUGGLE A JAPANESE PUZZLE BOX IN A WOK.

THE BOX IS AJAR!

QUIVERING IN ANTICIPATION, I QUICKLY WRITE A BANANAGRAMS ODE.

"FEED LEFTOVER WORDS

TO HER VULTURE.

DECODE THE MYSTERY FOOD."

---

HOWTO:
The process of writing such a thing is not unlike playing a game of Bananagrams: it's easy at first, but then as you approach the end, you realize that you have to rearrange and optimize things more carefully and then iterate. "Iterating, I converge on a solution. Bananas!" Depending on how picky you are, it can take a really long time.

I didn't use this, but a useful guide to this kind of project is the The Art of Long Anagramming page.

If you would like to make your own Bananagrams poem, pangram, or other long anagram, this site has a Flash application that helps you track the number of each letter and the total sentence length. There are also downloadable versions for offline use.

Finally, this anagram generator has an advanced search, allowing the requirement or exclusion of words which is incredibly helpful.

Hypothetical huge word grid

I heard about Monopoly City Streets, the online Monopoly game which uses a world map as the game board and streets all over the world for properties. It made me wonder: What would a Bananagrams grid that incorporated all the words in the dictionary look like?

If we use the Scrabble Tournament Word List (TWL), we get
101 + 1015 + 4030 + 8938 + 15788 + 24029 + 29766 + 29150 + 22326 + 16165 + 11417 + 7750 + 5059 + 3157 = 178,691 words. (This list neglects words longer than 15 letters, since they can't fit on a Scrabble board.) If we assume that every word intersects with two other words, this will require about 1,227,094 letters.

Each Bananagrams tile is a square, 1.9 cm (0.75 inches) on a side, so with an average grid density of 0.63 (my rough estimate), a Bananagrams grid with all the allowed words would be a square, about 1400 letters on a side, measuring 27 meters (87 feet) each way. That would cover an area a bit larger than one-and-a-half full-sized basketball courts.

How long would it take to assemble? If you suppose that a solution has been computer-generated and turned into instructions, someone who could place 10 words per minute, working 12 hours per day, would require 25 days to put it together.

Bananagrams Jargon

Most of the terms used when playing Bananagrams ("split", "bunch", "peel", and "bananas!") are all banana-themed. "Dump" is an obvious outlier - it doesn't quite fit. Some alternatives I have heard suggested are: "slip", "bad banana", "rotten banana", "bad spot". "Slip" has the one-syllable advantage, but sounds a bit too much like "split" for my taste. Maybe an exclmation like "Yuck!" would work. The "bad/overripe" theme seems appropriate, but of course "Rotten Banana" is reserved in the official instructions for someone who claims to have won but turns out to have used one or more bogus words. The word for the little gray bit at the end of the banana might work, assuming it has a name...

How about those stringy things on bananas? They are called "phloem bundles". They distribute nutrients throughout the banana. They are technically edible, but no one likes them, and people often throw them out. The analogy is perfect, and it's a short word (pronounced /FLOW em/). (Incidentally, peeling the banana from the other end (as monkeys do) is said to reduce the phloem problem, as the strings tend to stick to the peel.) The next time I elect to dump a tile, I'm going to say "Phloem!" and see how the other players react.

One final possibility (as a house rule) is simply not saying anything at all. People are generally too engrossed in the game to care.

History of anagrams: Anagrams as encryption

Galileo was a professor in Padua, Italy, studying physics and mathematics, when he heard about the sudden development of telescopes in Holland. Guessing at how they worked, he created his own in 1609 and began promoting it for the monitoring of ships by merchants.
By the fall, he began using it for astronomical observations. He designed a series of successively better telescopes, giving him the ability to discern smaller details of the night sky than anyone before him. This opened up a gold rush of scientific exploration. Galileo was able to make astronomical discoveries very fast, but such discoveries could sometimes take months or years to verify, which left him with a dilemma: If he announced his discovery immediately, he could eventually turn out to have made a mistake; if he waited, one of his competitors might find the same thing and get the credit. Galileo's solution (and possibly one already in use by other scientists) was to write a short description of his findings, rearrange the letters, and distribute the coded version. In that way, he could reveal his conclusions at any time in the future.

In 1610, he sent letters to his fellow scientists, containing the following string of letters:

smaismrmilmepoetaleumibunenugttauiras

Fellow astronomer, Johannes Kepler anagrammed very hard and came up with this solution:

Salve umbistineum geminatum Martia proles.

which translates to "Be greeted, double knob, offspring of Mars", and which Kepler interpreted to mean that Galileo had discovered that Mars has two moons, something that Kepler had predicted. But it turned out that Kepler's anagram was not the one that Galileo had had in mind. Allegedy the Holy Roman Emperor became interested in Galilelo's finding, and so Galileo finally revealed the actual original sentence to be

Altissimum planetam tergeminum observavi.

which translates roughly to "I have observed the most distant of planets (Saturn) to have a triple form" (where U and V are treated as interchangeable in Latin). Galileo thought that he had observed two moons orbiting Saturn. He was wrong, as Huygens showed in 1656 that what Galileo had seen was actually a ring around Saturn. Huygens wrote an anagram about this, too.

In December of 1610, Galileo made an even bigger discovery which he transmitted as a true anagram:

Haec immatura a me iam frustra leguntur oy.

which, translated, means roughly "This was already tried by me in vain prematurely". Kepler pleaded with him to reveal what he had found, and so in January, Galileo replied with the unanagrammed sentence:

Cynthiae figuras aemulatur Mater Amorum.

Translation: "The Mother of Loves [Venus] imitates the figures of Cynthia [the Moon]." by which Galileo meant that Venus cycles through phases, just like the Moon. He had observed that Venus is not a light source - it simply reflects light from the Sun. And from the way it waxed and waned like the Moon, he could tell that Venus was orbiting the Sun. At a time when many held that everything revolved around the Earth, this was an amazing result.

In the 1670s, Robert Hooke was studying the physics of springs. He found that when you stretch a spring, the force that it pulls back with is proportional to the distance that you pull it. This came to be known as Hooke's Law. Hooke encrypted this in a popular way among scientists of the time: He simply alphabetized the letters and got "ceiiinosssttuv". The unscrambled version is "Ut tensio sic vis." which he revealed a few years later to mean "As the extension, so the force.".

Hooke held on to another discovery much longer. In 1671, he announced to the Royal Society of London (a group of scientists, dedicated to advancing scientific knowledge) that he had figured out what the optimal shape for the arch of a bridge was. Four years later, he published the encoded form of his conclusion:
The unanagrammed form was not revealed until after he died in 1703(!). It read "As hangs a flexible cable, so inverted, stand the touching pieces of an arch." which means that if you hang a chain between two poles, you get a special curve called a catenary, and if you turn it upside down, this is the best shape for supporting a bridge.

I am guessing that the practice of anagramming conclusions in this way gradually disappeared as more formal publication methods were developed (the earliest probably being the journals published by the Royal Society of London starting in 1665).

In a way, this kind of encryption was not so different from the seeking of metaphysical truth through anagramming (as covered in a previous post), except that the scientists were deliberately putting the distilled form of the truth on the other end of the anagram and letting people search for it.


For more background on Galileo and his anagrams, you can read http://www.mathpages.com/home/kmath151.htm.

For more about Hooke's law and his caternary findings, see: http://www.lindahall.org/events_exhib/exhibit/exhibits/civil/design.shtml.

And for a fictionalized account of 17th century science and the development of the Royal Society (among other things), try Neal Stephenson's Baroque Cycle.

An event for people who love puzzles

A very unique event is happening in a couple of weeks: On September 13th, a puzzle hunt (a series of puzzles, spread around a particular area) will be simulcast to a bunch of different cities, including: Boston, Houston, Los Angeles, San Francisco, Portland, Seattle, and Washington, DC. The hunt is called DASH which stands for Different Area, Same Hunt. This is entirely run by volunteers and is an effort to expand the puzzle hunt community beyond its current ecological niches in the Bay Area and Seattle.

It will not be all words games, but if you are good at Bananagrams and like puzzles, you might enjoy it. If the sample puzzles appeal to you, I would encourage you to get together a few friends and give it a try.

UPDATE: DASH 2 is happening April 24, 2010.
2011 UPDATE: DASH 3 takes place April 30, 2011.

History of anagrams: Anagrams as prophecy

One half of the Bananagrams family tree was sketched out in this post which basically covered the inheritance of traits from crossword puzzles by way of Scrabble. But the inventor of Scrabble had another influence beyond word games that arranged words in intersecting columns and rows: An emphasis on anagramming. So, I've dug into the historical roots of anagrams. Here is some of what I have found:

While evidence from the time of the Greeks is spotty, it is thought that Pythagoras and his Pythagoreans may have been rearranging the letters of words in the 6th century B.C., as they believed that anagramming someone's name could reveal information about their destiny. Later, in 332 B.C., Alexander the Great was in the midst of a six-month siege of the city of Tyre, when one night he had a nightmare about a satyr trying to catch him. He asked his soothsayers about it, and they noted that in Greek the word for "satyr" anagrams to "Tyre is thine.". The next day, Alexander finally captured Tyre.

From a very insightful book called The Puzzle Instinct: The Meaning of Puzzles in Human Life by Marcel Danesi:

Soothsayer status was, in fact, often attained by those who claimed to possess knowledge of anagrams. In the third century B.C., for instance, the Greek poet and prophet Lycophron made a profession of devising anagrams of the names of the members of the Hellenistic king Ptolemy II's court in Egypt, as a basis for divining each persons' character and destiny. For this he became widely known and sought out as a soothsayer.

After the time of the Romans, there is little evidence of anagram use until its resurgence in the Middle Ages. As most educated people were associated with the Church, most anagrams from that era relate to religion and are in Latin. The most popular example is this imagined exchange between Pontius Pilate and Jesus:

Pilate: Quid est veritas? ("What is truth?")

Jesus: Est vir qui adest. ("It is the man before you.")

I have to admit, it would be really cool to answer somebody with an anagram of their own question.

The Kabbalists were mystics from the Middle Ages for whom anangramming held particular importance. To understand their interest in manipulating letters and numbers, it helps to know a bit about the ancient Hebrew language.

Ancient Hebrew did not have a separate system for numbers, so the letters of the Hebrew alphabet were used instead. This led to numerology (adding up the numerical values of letters in a word, and regarding the sum as significant to the word). Ancient Hebrew also did not have a conventional alphabet; it had an "abjad" which is a writing system where vowels are not written, but inferred by the reader. This made anagramming much easier. (Just imagine how much easier it would be to play Bananagrams if you could use "NRG" for "energy", "GRN" for "green", and "RNG" for "ring"!)

For the Kabbalists, the letters of the Hebrew alphabet held special powers. They developed a wide variety of rule-based systems for transforming one word into another. Some, like Atbash, were simple substitution cyphers (A->Z, B->Y, C->X,...). Temurah (the Hebrew word for "permutation") was the Kabbalistic art of anagramming. The Kabbalists believed that through such manipulation of names and verses in the Torah, they could uncover fundamental knowledge about the universe. They wore anagrams on amulets which they believed protected them from evil (a practice common among many others in the ancient world)

My guess is that if they had had the opportunity to play, the Kabbalists would have been Bananagrams maniacs.

More anagramming history, coming in a future post.

[Meta] A new domain name

I have set up this blog with its own domain name: bananagrammer.com. It's still hosted on Blogspot for now, so the old URLs should still all work. Specifically, if you use an URL that looks like bananagrammer.blogspot.com, Blogger will automatically forward it to www.bananagrammer.com.

If you are subscribed to the feed, you might have to change the feed URL. Or you can just delete the old one and resubscribe through the new feed.

For the moment, everything else is the same as it was, but in the future, I will have the flexibility to try out things that do not fit neatly into a blog post.

So I played the online version of Bananagrams...

I finally played the Facebook version of Bananagrams. (I suspect that the other versions (Bebo and MySpace) are basically the same.) It is a Flash application with a colorful interface and sound effects.

At first I trained on "Solo Cafe", a timed solitaire game where all the letters were revealed at the beginning. The first game I got an easy set of letters and was basically struggling with the mouse controls. I then played a few more games, doing a bit worse. Finally I was able to beat my initial time (by about 50%), so I figured maybe I was ready to try playing against live opponents.

I was faced with a screen showing many games in progress and one that was waiting for enough players to start. I clicked on the "waiting" one, which may have been a mistake because as soon as it came up on my screen the game started. I think there were four other people in the game at the beginning. I didn't have much time to look at what was going on in the sidebar (where you can see pictures of your opponents and very tiny depictions of their live grids). I did notice when one guy exited the game as there was some kind of brief pop-up notification. (Seeing the grids and remaining letters of one's opponents in a heads-up display would be a nice option for meatspace Bananagrams.) I was the first to finish with all of my letters, but then I realized that the game wasn't over. (Playing Solo Cafe had acclimated me to not peeling.) So I clicked on the pile of tiles in the upper left corner to start peeling. About five peels in, other players caught up. As my surplus letters piled up, I found a way to form them into words, and so I was the last to peel, leaving just two tiles in the bunch. I found a place for the last tile and clicked on the just-materialized banana icon in triumph.

The challenge is adjusting to the mouse control. As far as I can tell, there is no keyboard control. You can click on the letter you want and drag the tile to where you want it. You can also select many tiles by clicking on the table and dragging to form a selection rectangle.

You can then drag the group of tiles together. Also, immediately after selection a circle with a curved arrow in it appears near the group of letters. If you click on the circle, the group of tiles will rotate 90 degrees clockwise, like this:

Once learned, these tools would be useful when major grid-restructuring is necessary. Elite online Bananagrams players are probably separated by subtle mastery of the interface and selection/rotation tricks.

I can imagine that implementing the controls differently would make the game much easier. Imagine, for instance, that you click with the mouse where you want to place a letter and then type the letter that you want there. And then if you type the next letter in the word, it should automatically be placed, too.

From posts about surveys on gameplay, I suspect that the interface will be improving in the future. Aside from the interface issues, online Bananagrams resembles offline Bananagrams in one important regard: it is a addictive and quite fun.

---

Further reading: You might also enjoy reading my review of an online speed Scrabble game.

Also: tips on how to improve your Bananagrams performance and the longest word that you can make in Bananagrams.

Bananagrams is coming in new languages and for new platforms

A recent news article gives an overview of Bananagrams and describes many recent developments. Among the subtler changes, they now list as one of the official Bananagrams variations (like Banana Solitaire and Cafe), "Banana Challenge", which is just regular Bananagrams with the no-two-letter-words rule. In addition to the existing English and Spanish versions, Bananagrams translations are being developed for French, German, Italian, Norwegian, and Hebrew. Board games are apparently big among Israelis:

...Bananagrams is popular with Hasidic Jews because they are unable to use numbers on the Sabbath. Since the game only utilizes letters, they can play without violating the Sabbath.

The article goes on to say that electronic versions of Bananagrams are in the works for the Wii, the Playstation, and "Gamebox" (which I am taking to be a mistaken transcription of either Game Boy or Xbox).

I'm very curious to see what the Wii interface for Bananagrams will be like.

UPDATE: French, Norwegian, German, and Italian Bananagrams are now available in their respective countries.

And (at last) you can buy the French version of Bananagrams through Amazon in the U.S.!:


I just learned that the Norwegian edition of Bananagrams has a product page on Amazon. But sadly, it is listed as "currently unavailable". Apparently, the additional Norwegian letters are Æ, Ø and Å.

Hebrew Bananagrams is still not available... For now, the closest approximation would be buying one of these kind of expensive Hebrew Scrabble sets and playing almost-Hebrew-Bananagrams. Shalom!
And Hebrew Bananagrams is now available, too.

What to do when you have too many vowels (other than panic)

This is the natural complement to the post about what to do when you have too many consonants. As mentioned before, if you dump one of your vowels, assuming that you are picking from the 144-letter Bananagrams distribution, you have an 8% chance of getting three vowels back. In my experience, if you have one extra vowel tile, and you just can't rearrange the grid to fit it in somewhere, even exchanging it for two vowels and a consonant tends to be easier to deal with. Of course, you could wind up with some consonants that are tricky to use (which may be why I rarely dump letters.. I also enjoy rearranging the grid).

Whatever approach you choose, you definitely want to know some words with a large percentage of vowels. There are some nice all-vowel words out there like aye, eye, and you. Here is a sampling of some longer vowel-heavy words:

eunoia (83% vowels)
eerie, adieu, audio, bayou (80%)
year, ooze, area, iota, auto (75%)
sequoia (71%)

"Eunoia" and "sequoia" are also distinctive for being two of the shortest word containing all five vowels. "Eunoia" may get you in trouble if you try to use it since it's an obscure word. [I am partial to it because it is a brain word. It comes from a Greek word meaning "beautiful/favorable thinking". The "beautiful thinking" interpretation led to the obscure English usage of "eunoia" - a state of normal mental health. A stricter reading suggests that the Greek word referred to thinking that was favorable to someone (like one's spouse). The "blissful and benevolent state of mind" interpretation, though questionable, is the nicest.]

If you are interested in obscure words on the extremes of human language, check out the All-Vowel Words and All-Consonant Words dictionaries. They start with tame words like "eau" and "brr" and then spin off into highly arcane references (at times approaching Borges-level bizarreness). They come packaged together in a book called "Wye's Dictionary of Improbable Words", downloadable from Lulu for ~$14 or buyable from Amazon for $25.

Pangrams and what to do when you have too many consonants

Pangrams are sentences that use every letter in the alphabet at least once. Just a few examples:

• Nth quark biz gyps cwm fjeld vox. (which sounds like the shortest science fiction story ever)
• Vext cwm fly zing jabs Kurd qoph.
  which I like for the description:
  "An annoyed fly in a valley, humming shrilly, pokes at the nineteenth letter of the Hebrew alphabet that was drawn by a Kurd." (though I prefer the variation: Zing! Vext cwm fly jabs Kurd qoph.)
• Squdgy fez, blank jimp crwth vox! (Translation: An imperative sentence, commanding one's squashed-down brimless hat to mute the skimpy voice of a Celtic violin. I like this one particularly, as it was made by my current hero, Claude Shannon, whom I will write about eventually, elsewhere.)

There are many more 26-letter pangrams. By adding two letters, one can make parsable pangrammatic sentences like:

• Waltz, bad nymph, for quick jigs vex!

As pointed out by Robert Munafo: "The fact that so many optimal pangrams exist in English results mainly from its highly complex and flexible grammar, and the borrowing of so many words from other languages."

In particular, all of the 26-letter pangrams above use Welsh words (which have made their way into the English language), since in Welsh, the letter W is a semivowel. These words are a great way of dealing with a consonant-heavy round of Bananagrams. Check them out:

• cwm (pronounced /koom/) - a Welsh word meaning circular or bowl-like valley, often formed by a glacier.
• crwth (/krooth/) - an archaic Welsh stringed instrument - like a boxy violin (Parenthetically, writing this definition has inspired me to make up my own (suboptimal) pangram: The King of Quips seized the crwth (like a boxy violin) to jam.)

Another one is bwlch which means mountain-pass, and is kind of fun to say. "I pity the /boolch/!".

Some other good words for getting rid of consonants:

• hymns
• fjords
• rhythms
• strengths (the longest dictionary word with only one vowel)

The other option when you have too many consonants is, of course, to dump one of them, in hopes of getting some vowels. There are 63 vowels in a Bananagrams set (counting Y as a vowel), so if you pick 3 tiles from a random pool (starting with all 144 tiles), the probability of getting all consonants would be 81/144*80/143*79/142 which is about 18%. (The probability of getting all vowels is 8%.) [The odds could be computed more efficiently if you took into account all the tiles already turned over. This could be useful near the end of the game, but would seem to require superhuman ability to catalog all the displayed tiles. It seems more plausible in the final move of a Scrabble game.]

So, about 80% of the time you'd probably come out ahead (with one or more extra vowels) though possibly with much rearranging to do to use your new tiles. The other 20% of the time you'd just have more consonants and be even more stuck! This is one reason that I generally only dump at the beginning of a game or else when absolutely necessary.

Letter distributions in Bananagrams and other games

What determines how many of each letter there are in various word games? The inventor of Scrabble designed the frequency of his tiles around the frequency of letters used on the front page of the New York Times. (Somebody must have sat and done a lot of counting back in 1938.) The distribution for Snatch is said to be based on the frequency distribution of letters in the English language as calculated by World War II cryptographers.

This photograph visually displays the letter distribution for Bananagrams. (Click for a larger version.)
The following table gives the letter distributions for Bananagrams, Scrabble (without the blanks), and Snatch:

Game	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O	P	Q	R	S	T	U	V	W	X	Y	Z
Bananag.	13	3	3	6	18	3	4	3	12	2	2	5	3	8	11	3	2	9	6	9	6	3	3	2	3	2
Scrabble	9	2	2	4	12	2	3	2	9	1	1	4	2	6	8	2	1	6	4	6	4	2	2	1	2	1
Snatch	5	2	4	4	12	4	2	5	5	1	1	5	4	5	6	3	1	5	5	7	4	2	3	1	3	1

By comparing the Bananagrams and Scrabble sets, we learn that the Bananagrams set of tiles is a superset of the Scrabble set of tiles (except for the blanks). In fact, for each letter of the alphabet, the Bananagrams set has at least one more tile than the Scrabble set. Since there are 100 Scrabble tiles and 144 Bananagrams tiles, this additional 26 tiles explains a lot of the discrepancy. It also gives a hint as to how the inventors of Bananagrams may have derived their distribution.


The Snatch set, on the other hand, is NOT a subset of the Bananagrams set. If you want to try to play Snatch with a Bananagrams set of tiles (and if you are kind of a stickler), you can get close to the right distribution by throwing out these letters in these numbers: (A,7), (E,5), (I,6), (O,4), (R,3). (I am reducing each difference by one to allow for a slightly larger Bananasnatch set (121 tiles instead of 100).) You can remember this with the mnemonic lAttIcEwORk and noting that you start by throwing out 7 tiles lettered A, decreasing the number by one for each of the vowels, and finally setting aside 3 tiles with the letter R on them.

UPDATE:
I found the letter distribution for Spanish Bananagrams here and the French Bananagrams letter distribution here.

Here is the revised table, in vertical format (to accommodate the extra letters):

	B a n a n a g r a m s	S p a n i s h B G	F r e n c h B G	S c r a b b l e	S n a t c h
A	13	17	13	9	5
B	3	3	3	2	2
C	3	5	4	2	4
D	6	7	6	4	4
E	18	17	18	12	12
F	3	2	3	2	4
G	4	3	2	3	2
H	3	3	2	2	5
I	12	8	12	9	5
J	2	2	2	1	1
K	2	1	1	1	1
L	5	5	6	4	5
M	3	3	4	2	4
N	8	7	9	6	5
O	11	12	12	8	6
P	3	3	4	2	3
Q	2	2	2	1	1
R	9	7	10	6	5
S	6	8	6	4	5
T	9	6	10	6	7
U	6	7	6	4	4
V	3	2	4	2	2
W	3	1	1	2	3
X	2	1	2	1	1
Y	3	2	1	2	3
Z	2	2	1	1	1
CH		2
LL		2
Ñ		2
RR		2

---

Suggested reading:

• The Boggle cube redesign and its effect on the difficulty of Boggle
• The longest word that you can make in Bananagrams

The game of Snatch (a.k.a., Anagrams)

I recently learned a great game that can be played with the Bananagrams tiles. It is currently sold under the name "Snatch", but earlier versions actually pre-date Scrabble and, if the Wikipedia entry can be believed, has existed in some form since the Victorian era.

Like Bananagrams, it can be played with 2 to many players, but also like Bananagrams, it seems like (tentatively) 4 is the optimum number.

The object is simply to build words from the available tiles. Start with all the letters face down in the center of the table. One player starts turning over tiles until someone sees a combination of tiles that make a word (at least 3 letters long). They shout out the word and claim it. Then the next player starts turning over tiles. What makes the game interesting is that, even a word that have been claimed by a player can be snatched away if someone sees a way to combine all of its letters with one or more tiles from somewhere else on the table to make a new word. New words must have a different meaning. (You can't just use another form of the word. For instance, if someone has the word TONE, the variations TONES and TONED are not allowed. STONE, on the other hand, would be a legitimate snatch.) You can build off of your own words, and you can combine the letters from multiple words to form a single new one.

The rule that I really like is that, if you see a word, you can snatch it at any time. For example, Alice turns over a tile. Bob sees a word and snatches it. Your first reaction (from the paucity of options when the game first starts), is to wait for Bob to take his turn flipping tiles. But if you see a word, you can grab it immediately.

Ties: If Alice and Bob both call out words at the same time and the words share at least one tile, the conflict can be resolved by yielding to whoever has the longer word. You should make up your own house rule for when two players call out the same word (or words of equal length) at the same time. (One idea is to throw those tiles back into the pool.)

When all the tiles have been turned over, play continues until everyone decides that there are no more moves to make and quits. Scoring is completely arbitrary. I like counting the number of words, but admittedly, this penalizes for combining two of your own words to form one longer one.

You can play Snatch with Bananagrams tiles (which I call "Bananasnatch"), but you may find that there are a few too many vowels as the Bananagrams set is optimized for Bananagrams while the Snatch set is designed for playing Set. I threw out about 12 of the vowels and found that we had a pretty good game. Throwing out more may make the game steadily more challenging. I will have a more detailed analysis of the letter distributions in a forthcoming post.

After playing one round of Bananasnatch, everyone wanted to play again. It is tremendous fun! It's perfect for when you've played six or seven rounds of Bananagrams and you want to change things up a little bit.

Appletters & Pairs in Pears: New games from the makers of Bananagrams

I stumbled onto this blog entry, describing a bit of the Bananagrams origin story, and then revealing that the Bananagrams people are about to release two new games!

So far, I've only found out that Pairs in Pears has tiles that augment the letters with patterns (dots, lines, blank and solid), and that it is aimed at kids, helping them to work on "cognitive skills and memory, as well as the order of the alphabet, vocabulary and rhyming as they practice forming words".

And I can find no information at all about Appletters. Not even how to pronounce its name. Just that it exists. Here is a fuzzy picture of Appletters:

If you have any information about the nature or current whereabouts of either of these games, please send it in to our anonymous hotline.

UPDATE: I have obtained what appears to be an official description of Appletters: "A fun domino game... You don't connect the dots -- you connect the letters!"

UPDATE^2.1: I've posted new information about Appletters here and Pairs in Pears there.

Bananagrams online tournaments announced

According to the Bananagrams Twitter feed, there will be a tournament for the online version of Bananagrams (i.e., through Facebook and possibly other sites) on Monday, July 6th, 2009.

The event is described here (Facebook login required, but here is the relevant text) :

This is it, folks...the moment you've been waiting for.

You all have been practicing, so now it's time to show your stuff! The Bananagrams "Happy Monday" contest is a way for YOU, the players, to earn prizes and gain fame!

Here are the rules: Beat as many players as you can in Live Multiplayer games on Monday, July 6th, from 12:00am-5:00pm. That's it. Simple, right?

* GRAND PRIZE: 5,000 free Benjees and a set of elite TROPHY TILES (not available in the store).

* 2nd place: 2,500 free Benjees and 10 Charms.

* 3rd place: 1,000 free Benjees and 5 charms

* All top 25 players receive 250 Benjees and 3 charms.


The excitement begins on Monday, so start practicing and get ready to win! GOOD LUCK!

It's a virtual game, so all the prizes are virtual, too. I'm guessing that the "Benjee" is the online Bananagrams unit of currency for buying things in the virtual Bananagrams store. I do like the look of those trophy tiles!

UPDATE: This has turned into a regular, weekly event.

LATER UPDATE: These online tournaments have ceased, but I'm leaving this up for historical purposes.

"Bananagrams!: the official book" - coming in September, 2009

The first Bananagrams-themed book is on its way:
Bananagrams!: The Official Book. It sounds like the book has a huge variety of Bananagrams-based puzzles (presumably puzzles you can solve using your Bananagrams tiles) as well as tactics for Bananagrams and random fun.

About the authors:
Joe Edley, the Zen master of competitive Scrabble, was featured as one of the main characters in Stefan Fatsis' _Word Freak_, the original investigation of the world of Scrabble tournaments (followed up by two documentaries (Word Wars and Scrabylon)). I really enjoyed _Word Freak_.

And from the Amazon page:

Abe Nathanson and his daughter, Rena, along with his grandchildren Aaron and Ava invented Bananagrams while spending the summer of 2005 together at a beach house in Narragansett, Rhode Island. The whole Nathanson family is involved in the growing game business. Rena and her family reside in the U.K., and Abe lives in Providence, Rhode Island, where he runs the company's headquarters.

According to the official Bananagrams origins story, Bananagrams was created by the Nathanson family during summer vacations, through a process of experimentation. Not unlike the Wright Brothers' persistent tinkering until they achieved perfection.

Why "Bananagrams"? Because they say that it is "the anagram game that will drive you bananas!".

The ancestry of Bananagrams

What if a game like Bananagrams had been invented before Scrabble? Would Scrabble be as popular as it is today? Would it even have been manufactured? There are definitely people who like Scrabble but do not enjoy the pace of Bananagrams. But I would guess that Bananagrams could not have become popular before Scrabble, because the idea of the freefrom grid would have been a little too radical.

Scrabble was invented in 1931, and the inventor was inspired by crossword puzzles (which were a massive fad during the Twenties... picture women wearing crossword-puzzle stockings). And the inventor of the crossword puzzle was, in turn, inspired by word square puzzles where the objective was to construct a grid of tightly interlocking words to form a square (or other shape) based on clued words. A typical word square looks like this:

     C A R D
     A R E A
     R E A R
     D A R T

It is an interesting exercise to try to construct a word square,
By specifying the letters in such a square, this can be turned into a relatively easy puzzle.


Word squares as puzzles date back centuries, but the word square formation has been in existence for at least two millennia. (See the palindromic SATOR square.) Finally, the Greeks are credited with the idea of ordering letters in lined-up rows and columns (c. 600 B.C.), though without any words running vertically.

These are some of the giants whose shoulders Bananagrams perches upon. Remember to thank them the next time you pray to the Bananagrams gods for good tiles.

---

Further reading: The Straight Dope on why crossword puzzles are symmetrical.