Released May 14, 2022.
My memory technique involves posting images along a predetermined imaginary route as is the common approach that mnemonics generally use. The names made up for these images are encoded with numbers. Many mnemonics build their own variation of the Major System, a system invented centuries ago. Here's my own variation of the basic Major System for the Word Structure Memory Method:
0- S, Z, C (soft sound)
1- T, D, Th
3- M, W
4- R, H
5- L, X
6- J, Sh, Ch (soft sound), Ng
7- K, C (hard sound), Qu, Ch (hard sound)
8- F, Ph, V and Y (as in 'yes')
9- P, B
My key includes a greater variety of sounds than you find in the regular or basic Major System. I welcome more sounds so that any word can easily be turned into a number.
0 = U (either the English or french sounding 'u') first, 'ou' (french sound) secondary choice. The English 'oo' sound as in 'book'might be given this values too but given the two 'o's, it may also work as a #2 and will not be considered a dominant sound in whatever category.
1 = A (dominant sound as in 'tack'), then 'a' as in 'art', and lastly is the sound 'ay' as in 'pay'.
2 = O, OA (dominant sound first, such as in boat) then 'on’ (french sound as in 'pont'), then 'o' as in boss.
3 = EE, (dominant sound as in pee) then a soft 'e' (french or english sound as in 'best', and also a dominant french 'e' sound as in 'été')
4 = same as #0
5 = same as #3
6 = i (English dominant sound as in 'pi')
7 = same as #1
8 = i (non-dominant, such as 'pit') french 'in' sound as in 'peintre'
9 = same as 8
The use of 2 identical sounds to identify a number is a rather novel approach. It means that there will be a slight possibility of error in decoding the words but I think the benefits of going that way outweigh the inconvenience. You get to realize this as you use the system.
The time has now come to start explaining the truly revolutionary aspect of this novel Word Structure Memory Method: I am simply saying, let's be more creative: Let's assess the few various structures normally found in words with in mind to use that as a hint or key telling us which coding or decoding method to use to come up with its 3 digit value.
What follows are all the key structures of words that we will use to assign numerical values to words in a more systematic and practical manner:
First, let's look at some repetitive numbers such as 333, or 888. Under the traditional Major System these numbers for words tend to be challenging to find an associated image as few words tend to repeat the same type of consonants 3 times in a row. But structurally, these numbers repeat themselves and so they stand out memorably. So for such numbers, it's fitting to match them to an abbreviated structure: Simply use a one consonant and one vowel word, such as ma, or me. In the Major Sytem, 'm' is valued at 3. So, for example, 333 is a special format number and I may use 'me', (one's self image) to represent it.
When it's time to decode the 'me' image into a 3 digit number, you immediately notice that the word is noticeably short and the reason for this is that it represents a number without much variety to it, such as a repeating number and the consonant gives you the number that is being repeated by using the key in the Major System.
Meanwhile, the actual simple digit 3 ends up being memorized as number 003 rather than just a one 3 standing alone. Structurally, 003 and all other 3 digit numbers that have in them a doubling of digits next to each other are special as they are not completely made of non-repeating digits. What word structure can we use to reveal this? Well, let's use this format: the first letter of the word will be a vowel then next to it will be a consonant then vowel then consonant at the end. So 'assume' or the french word 'assome' would work there to represent 003.
Here is the key: The presence of an initial or end vowel indicate a repeat of the value of the adjacent consonant. For the number 033, you could use the name Zoomy (the nickname of memory champion Ben). The 'y' at the end is pronounced as a vowel sound 'ee' and so it indicates a repeat of the vowel next to it the 'm', doubling the 3 at the end. So although the presence of an initial or end vowel in a word is normally indicative of a digit (its own intrinsic value from the Major System) that value isn't used because the word structure gives you a hint to come up with a value for the word in a different way. As the word structure appears to lean with more sound weight to the left (beginning) or right (end) with a vowel present there, it seems fitting that it would indicate a double load of a digit at that spot.
Now you may wonder why that is being done and not more simply just use the intrinsic value of the vowel from your system of assigned numbers to vowel. The reason is that this new system has greater flexibility to match more words conveniently and you quickly realize this as you work to select your images using this system.
For more variety and more choices, I have observed that it's not always useful to always double up the last or first consonant with an end or beginning vowel: When the structure of the word differs, you should give yourself greater choice. For instance, let's take the word 'plow'. As you may notice, there are no vowel in between the 'p' and the 'l' and that's your structural signal indicator telling you that the value of the last consonant will not be repeated as it was with the word 'Zoomy' which has a big 'oo' vowel sound between the 'Z' and 'm'. So what do we do then? Well, my suggestion is to take the value that you assign to the last vowel (from the above vowel key) and use that instead, so plow would become 952 instead of 955 as my value for the vowel 'o' in my system is 2.
Let's keep going. If any word has 3 major consonants in it then these 3 consonants' value need to be used. (That's called the Major System and it's not new.) If there are more than 3 consonants, I suggest using the first and last 2 consonants as we often remember best the first and last part of anything we memorize.
When the word structure you have for your image is made up of a single consonant and it is sandwiched by 2 vowels, such as 'Eva', I suggest assigning that structure to numbers that also have a sandwiched appearance. For instance, 878 could be Eva, where the middle consonant's value is simply duplicated 2X and relocated at the number's extremities. Provided that at least one of the vowels of the sandwiched word has the same value as the center digit this word becomes easily memorized to its number. So when you find an ivy plant in your memory castles, the word structure of 'ivy' tells you to double up the value of the center 'v' to get a sandwiched looking number and use the value of either vowel for the center digit. I use 'ivy' as an 868 number.
With practice, another basic word structure appeared useable: for instance one consonant followed by two vowels such as Zoe. In that case I will use each letter's value in the order they are. For instance 'Zoe' turns into 023.
There is finally one more word structure to discuss, the one consonant and one vowel structure. I talked first about it, if you remember, for triple repeats of digits.
Since there are only a few numbers that are triple repeats of digits (just ten in a thousand), I think they can be memorized apart and we can still use another deciphering approach for that simple word structure. That's very important because there are so many words to choose from among that word structure. And being short words, it may help reduce the data load that mnemonics need to memorize. That last trick took me the longest to devise and get familiar with but I definitely vouch for it's worthiness. Here it is:
When a one vowel word starts and ends immediately with one consonant, you just take the value of the first vowel then the value of the consonant and then you just repeat that last value with one added. So, the word 'ash' for instance would become 167. On the other hand, if the word starts with the consonant and finishes with the vowel, you substract one from the last value. So, the word bee could become 932. This system is arbitrary and built for simplicity.
It is possible to imagine other approaches that would work quite well, and in retrospect, maybe even easier to memorize then what I have been training with: you may for instance just follow the vowel (toss the word left or right so that it always has the vowel centered on a word's edge -in the 3 digit empty grid- and fill the remaining void spot -either at the beginning or end- with the value of the center consonant minus or plus one). And the alternate way with the consonant on the edges is certainly possible too. Just be consistent with the system you adopt. This paragraph may need more details for easy comprehension of these other variations but I just wanted to throw in the notion that other approaches are possible and so I will leave it as is.
If your word is made of a begining vowel, 2 consonants and an end vowel, then you may want to use the value of the two consonants plus either one of the vowel values, preferably in order. So a word like 'acne' may be determined to be worth either 172 or 723. You just make the decision based on what you think will work best for your system. (Here too, you may twinkle the rule and use the value for both initial and end vowels and pick the value of a consonants for the middle digit. It may be an inferior/superior approach, as I haven't tried doing it that way but it might work for you.)
For the record, the basic word structure consonant - vowel- consonant will be read as a number using the respective values for each consonant and vowel found in the above keys. 'Bike' for instance, will be worth 967.
Interestingly, there is a different type of structural observation about words that may help you to number them better. Have you ever noticed when a vowel repeats itself prominently in a noticeable way? For instance, a 'tipi' would be worth 199 under my above rules but since the vowel 'i' -which is pronounced 'ee' in that word (well, at least in the french pronunciation)- is emphasized by being repeated twice, it can be used as the first number and 'tipi' can then be assigned the numby 319 instead. If you think (because your accent differs from mine) that 'tipi' is pronounced with a light 'i' then you would have the option of using it for numbers 819 or 919, according to my vowel key.
Let's cover some funny exceptions. A word like 'lily', according to the above set of rules would be worth 555 but 555 has been determined as a simple one consonant word per other simplification rules. So, in such cases, the obvious thing to do is not to repeat the last consonant but simply take the value of the last vowel, so 'lily' will be worth 553.
All of the above rules took me weeks if not months to develop so if you read this in a few minutes, it may seem overwhelming. But when you start building your own system with it, you end up feeling that this system is overwhelmingly practical. There is no going back to the original strict Major System for use with 1000 image systems or above.
There are lots of finer points or personal rules that you may find useful to you as you develop your images through practice and you will develop a sense of what is more easily memorable for you. You can bend the rules from time to time and doing such regular rule bending maybe more beneficial for some than for others. Let's cover a few of such rule bending instances:
As a kid, I watched Detective Colombo on TV. Now when I needed an image for 539 to go with 'lump' and 'lamb', 'lombo' as in 'Colombo' kept surfacing in my mind but 'lombo' is not a word. Limb or limp seem like obvious choices but they do not have all the stories that 'Colombo' evoke in my mind and so Detective Columbo became Lumbo (in a lambo/rghini). I prefer this also because of the dominant 'o's in 'Colombo' makes it easy to locate this number in my highest ranking thousands of images as 2539.
Inventing words may simply involve creating something new but almost identical to the original. For instance, the french word, 'érable' (a Maple tree) can be difficult to differentiate from the invented word, 'érab' as the 'b' and 'l' can melt together when pronounced quickly. The invented word would read as 449. Filling such number can be much more challenging than filling the 'érable'= 495 spot that can be better filled with more obvious suggestions such as 'rubble', 'Rebel', 'Rample’, etc.
How you go about developing your memory castle also matters: I keep records with the Anki application. I have 2 sets of 3000 cards. One is for the actual images I am using and the second set is my scrapbook of images. The scrapbook is very useful for when you need or want to change some image, or get a new idea for an image, or simply to keep a record of an image description. The other set is kept clean and is used to speed-read through to memorize the images.
The system you use to store your 3000 images in your head using memory castles is also critical to your ability to memorize all of this. However, there is nothing new to this part so I will simply recommend this link to learn more about the way that I strongly recommend that you go about memorizing all this imagery:
A memory castle built as I explain in that thread will allow you to easily memorize things using the Less l. Method, and will also make it easy to divide the images into two sets for use with the Block System invented by Johannes Mallow, if you prefer to train with fewer images. Doing it using these instructions will also help you realize why I have set the vowel key above the way that it is set.
Once you have your 3000 image set, you may further be able to refine your system by using a vowel or a group of vowels in each word to determine where to put your image, either in the 000s 1000s or 2000s. For instance, for 957, I use 'plug' for 1957, 'plaque' for 2957, and 'block' for 2957. That's because the 'u' is woth 0 or 4 in my system, the 'a' is worth 1 or 7 and the 'o' is worth 2. So up to 1000, many of my words will very often include an 'u' then up to 2000 it will be the 'a' that dominates and finally the 'o' is the vowel of choice up to 2999. If two words happen to have a similar vowels such as for my #218, 'native', 'Nadia', 'nettoyant: then 'Nadia', which has a more dominant sounding 'a' than the 'a' of native gets to be the one, 1218. Since 'nettoyant' has a 'o' it gets the level 2 or 2218. 'Native' may not have a 'u' in it but has no other definitive place where it should go and so it takes what ever spot is left and in this case, the 0 spot, or 0218. Because there is a connection between these 3 numbers, their respective images are best learned together. Because of that connectivity, I suspect that gaining a sharp recall for a 3000 image system may be achieved more easily than with a 1000 image system.
Let's try another example, 927, 'pock', 'barnique', 'poke'. Both 'pock' and 'poke' have an 'o' but the 'o' in 'poke' sounds more dominant and so it will attached to 2927. 'Pock' has an 'a' sounding 'o' but 'barnique' has a true dominant 'a' in it. To me, that's better and so 'barnique' will be be 1927 and 'pock' will take the only remaining spot, 0927. Sometimes when vowel differentiation isn't possible such as with words, 'Mick' and 'wig', I will then use the alphabetical order to order the words, 0397 for 'Mick', and 2397 for 'wig' (and use the 'm-back' for 1397, an invented word or nick name for a back flipping humpback whale)
That's it. There is some fine tuning that I could talk about but it would not be helpful since it might only help those who are advanced with this method and much of this fine tuning is best done by one's self.
Now let's talk about some of the advantages and benefits of building such a large system: Most people would believe that large systems form a bigger challenge to memorize and if instead you were to vigorously train using a smaller systems you would end up faster at memorizing much sooner. But on the other hand most memory champions use at least a 1000 image system for numbers and so big systems can work too. I will admit that a 3000 image system built as I suggest is not used by any top mnemonics right now but this may change as memory sports evolve rapidly.
Familiarity with two languages or more also greatly facilitates the task of building the system. If you don't speak a second language but would like to learn one this will be a great opportunity to help you learn new words in your new extra tongue. Building a large memory castle is an enjoyable activity. The images you choose end up forming your group of ambassadors that you will use to connect with data to memorize. When I look at flowers, I always look more attentively to see and recognize the ones represented in my systems, and so observing flowers or birds or vehicles becomes more stimulating.
With a 3000 image system, it's also quite easy to encode any two letters into an image and any four letter words can then be known by its root 2 images in your system, and these association will solidify the learning. If a four letter words is a name you are attempting to memorize but you're unfamiliar with it then your system can be amazingly helpful...And that's a clear advantage of my system over the Ben System for instance, (which has also more than 2500 images to memorize for a deck of cards). His system was designed to encode a choice of 13 ranks by 16 combo of 2 suits, which makes it very convenient with cards but not so much for regular numbers beyond 999. It doesn't have the versatility and has not been designed to be applied to numbers higher than a thousand and so it cannot be conveniently used to recall easily any two letter combination.
When you have memorized all the values for each letter, a to z, 1-26, the highest number you may get is 2626 for z-z. This is Josh in my system and again, there is no other memory system of this size out there that has such great coverage.
My system of 3000 images can also be used effectively as a 2 cards system: I simply ignore two face cards in the deck of 52 cards, assign the remaining 50 cards the values of 0 to 49 and again from 50 to 99. Since there is 60 groups of either 0 to 49 and 50 to 99 in the 3000 image system, it's easy to assign a specific and unique card to 50 of them and thus have the first card of your pair covered by the identify of the group in which your second card's image appears. When one of the missing cards comes up, I simply use the value of the previous card and when the other missing card comes, I use another system to memorize it, the Shaper or Rhyme Method which I find extremely useful to complement my method. It consists of memorizing 0-99 images using either the shape of the number or a rhyme with the number.
There is another aspects to this method that I believe makes it superior in training one's memory: because its use develops your conscious familiarity with the structure of a word, you will learn to pay more attention to this, and soon you'll find pleasure in translating every word into a number. That activity surely will help you to train your attention to a higher degree. You end up noticing the initials of people you know or meet to see what image that links to in your system... You also end up normally dealing with more data concentrated in smaller words. Learning to use words tightly packed with data to memorize is bound to make a positive difference in one's general memorization abilities.
Lastly, I want to mention that I have heard of a top memory champion who claims to use a 10,000 image system for numbers. That's not typical of what the majority of top mnemonics use today. If that champion were able to teach effectively how he does it, then his system would prove to be a better system but so far few (maybe none) students follow him down that path successfully. So his system may be considered an odd exception but still one well worth mentioning. It also shows that a 3000 or 3100 image system may not be that overwhelming to learn, especially if you apply this Word Structure Memory Method to build it.
Simon is the Chair of the Canadian Memory Championships and a mentor to new people interested in memory as a sport. Most Canadian Memory Champions have attended his free memory workshops at some point. Above is a photo of the participants in the 2018 Canadian Memory Championship.
Simon thinks that people in general could do more for the environment. And he believes one of the easiest thing to do for the environment is to go vegan. He also believes that it is important to live in harmony with other living beings and not to behave in ways that involve cruelty. Since there are rarely political candidates who support his views, he feels his best option is to run in elections with small parties who share his values. Simon was the first ever male candidate to run for the Animal Protection Party in 2008.
Simon is a former Toastmaster. He was the Queen's Park Public Speaking Club President for 2 years. He has achieved the Toastmasters Gold Communication Award. Under his leadership, the Club received top Awards from Toastmasters International
To pay the bills, Simon works as a courier. Early in 2013, the Toronto courier company he worked for, Pretty Darn Fast, decided to go out of business. He saw this as a business opportunity and started his own courier company, Simon The Courier, to take care of the clients that he was already serving anyway. Simon very much enjoys running his own business and looks forward to giving his downtown Toronto clients a great service! You can reach him at firstname.lastname@example.org.
Simon has worked on many campaigns with other activists. This includes supporting reforms for fairer elections or proportional representation. Most recently Simon worked on a local petition to protect the Double-Crested Cormorants in Ontario from a cull disguised as a new open season.
Simon also has a passion for politics for the environment and the animals
This site provides information about the Canadian Simon Luisi, who lives in Toronto. Site updated May 2022.