The “Shapely” Mnemonic System

The “Shapely” Mnemonic System

Over the past three years or so I’ve developed a straight-forward mnemonic system for numbers, and jury-rigged code to make it even easier to put in practice.  In the upcoming months I’ll look to add in the resources that make this quick and easy.  (Initially this will be in English only, sorry!)


The basic gist of the mnemonic system is pretty standard: like the Mnemonic Major system, we will replace every numeral by a letter, and construct words with those letters.  Let’s start with the code:

The shape similarity is what makes this easy to pick up, and why I’d call this the “Shapely” mnemonic system.  (I love toiling for elegant, simple solutions.)  Note that all of these are consonants – this is a very important feature for the next step.

To memorize a string of numbers we will construct words and phrases using the letters represented as anchors, spaced by every other non-represented letters.  For example, if we are to remember the digits 43364, we would construct a word which contains RMMbR, and any non-represented letters in between  them.  ReMeMbeR is one of these words, as is Row-MeMbeR, but RuNMeMbeR would not be valid since incorporating the N would suggest 423364.

And I hope you would agree that remembering the word “remember to water the monkeys” is much easier than the phone number 433.647.7473 (ext 25), and that given the shape-semblance, decoding isn’t hard at all.

Encoding, on the other hand, is more open-ended.  While Shapely has a great deal of flexibility, e.g., by not restricting the use of any vowels or common consonants like c, k and p, there is cognitive effort required in converting a number to a word/phrase.  A great deal of this ambiguity comes from what you’d consider a word.  For example, one of my friend lives at 1615 of a certain street, apt 224, and labless nanoreactor is what I, chemist, came up with and find incredibly memorable (note the advance construction of ignoring trailing letters – her building doesn’t have apt 22474, nor the street 16155!)  Most other people probably won’t think these are valid words to begin with.

Whatever the reason, straight-up encoding is taxing.  Wouldn’t it be great if this is automatic?  Concerned more with efficiency than individuality, I write a piece of code that scans a dictionary for words that fit the criteria.  In the upcoming months I’ll look to expand on the code to scan for probable combination of words for up to 8 digits, and embed a lookup form in this page where someone using the mnemonic can simply query what phrases are good to remember 47,191,859 with – preferably being able to choose one that can be combined with the spatially based mnemonic techniques.  Check back on the blog for the experimental progress (and inevitable set-backs).



[/one_half] [one_half_last] [button link=”” type=”big” newwindow=”yes”] Decoder[/button] [/one_half_last]

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.