Learn Any Language

("Iversen's method")

A word list technique is in its most common form a list of words in a target language with one translation of each word into another language, here called the base language. However, you can use short idiomatic word combinations instead of single words, or you can give more than one translation into the base language, and it will still be a word list. You can also add short morphological annotations, but there isn't room for examples or long comments in a typical word list. Lists of complete sentences with translations are not word lists.

There are also word lists with just one language (frequency lists) or with more than two languages. The so called Swadesh lists (named after Morris Swadesh) contain corresponding lexical items from a number of languages, typical 100 or 200 items chosen among the most common words. Both these lists can be valuable for a language learner who wants to make sure that s(he) covers the basic vocabulary of a target language.

Dictionaries can be seen as sophisticated word lists, where the target items (lexemes) are put in alphabetical order, and where the semantic span of each lexeme is illustrated through the use of multiple translations, explanations and examples, sometimes even quotes. In addition good dictionaries give morphological information about both the target language and the base language words. However the amount of information in dictionaries varies, and the most basic pocket dictionaries are hardly more than alphabetized word lists.

Using word lists[]

The most conspicuous use of word lists is the one in text books for language learners, where the new words in each lesson are summarized with their translations. However they are also an important element of language guides used by tourists who don't intend to learn the language of their destination, but who need to communicate with local people. In both cases the need to cover all possible meanings of each foreign word is minimized because only some of them are relevant in the context, - in contrast, a dictionary should ideally cover as much ground as possible because the context is unknown.

Using word lists outside those situations has been frowned upon for several reasons which will be discussed below. However they can be a valuable tool in the acquisition of vocabulary, together with other systems such as flash cards. The method that is described below was introduced by Iversen in the how-to-learn-all-languages forum as a refinement of the simple word lists, and it was invented because he found that simple word lists weren't effective when used in isolation (except for recuperation of half forgotten vocabulary).


[Super-fast vocabulary learning techniques, many pages]

[Iversen's thread in the forum ]


One basic tenet of the method is that words shouldn't be learnt one by one, but in blocks of 5-7 words. The reason is that being able to stop thinking about a word and yet being able to retrieve it later is an essential part of learning it, and therefore it should be trained already while learning the word in the first place. Normally people will learn a word and its translation by repetition: cheval horse, cheval horse, cheval horse... (or horse cheval cheval cheval cheval....), or maybe they will try to use puns or visual imagery to remember it. These techniques are still the ones to use with each word pair, but the new thing is the requirement that you learn a whole block of words in one go. The number seven has been chosen because most people have an immediate memory span of this size. However with a new language where you have problems even to pronounce the words or with very complicated words you may have to settle for 5 or even 4 words, - but not less than that.

Another basic tenet is that you should learn the target language words with their translations first, but immediately after you should practice the opposite connection: from base language to target language. And a third important tenet is that you MUST do at least one repetition round later, preferably more than one. Without this repetition your chances of keeping the words in your long time memory will be dramatically reduced.

This is the practical method: Take a sheet of paper and fold it once (a normal sheet of paper is too cumbersome, and besides you need too many words to fill it out). If you have a very small handwriting you can draw lines to divide it as shown as b) below, otherwise divide it into two columns as shown under a). The narrow columns are for repetition (see below). Lefthanders may invert the order of the columns if that feels more comfortable. Blue: target language, red: base language. Curvy top: original column, triangle top: repetition column Wordlist1.jpg

Now take 5-7 words from your source and write them under each other in the leftmost third of the left column. Don't write their translations yet, but use any method in your book to memorize the meanings of these 5-7 words (repetition, associations), - if you want to scribble something then use a separate sheet. Only write the translations when you are confident that you can write translations for all the words in one go. And use a different color for the translations because this will make it easier to take a selective glance at your lists later. If you do fail one item then look it up in your source, but wait as long as possible to write it down - postponement is part of the process that forces your brain to move the word into longterm memory.

OK, now study these words and make sure that you remember all the target language words that correspond to the translations. When you are confident that you know the original target words for every single translation you cover the target column and 'reconstruct' its content from the translations. Once again: If you do fail one item then look it up in your source, but wait as long as possible to write it down (for instance you could do it together with the next block) - the postponement is your guarantee that you can recall the word instead of just keeping it in your mind. So now you have three columns inside the leftmost column, and you are ready to proceed to the next block of 5-7 words. Continue this process until the column is full.

There isn't room for long expressions, but you can of course choose short word combinations instead of single words. It may also be worth adding a few morphological annotations, but this will vary with the language. For instance you could put a marker for femininum or neuter at the relevant nouns in a German wordlist, - but leave out masculinum because most nouns are masculine and you need only to mark those that aren't. Likewise it might be a good idea to indicate the consonant changes used for making aorists in Modern Greek, but only when they aren't self evident. In Russian you should always try to learn both the imperfective and the corresponding perfective verb while you are at it, and so forth. You can't and you shouldn't try to cram everything into your word lists, but try to find out what is really necessary and skip the details and the obvious.


You can get your words from several kinds of sources. When you are a newbie you will probably have to look up many words in anything you read in the target language. If you write down the words you look up then these informal notes could be an excellent source, - even more so because you have a context here, and it would be a reasonable assumption that words you already have met in your reading materials stand a good chance of turning up again and again in other texts. Later, when you already have learned a lot of words, you can try to use dictionaries as a source. This is not advisable for newbies because most of the unknown words for them just are meaningless noise, but when you already know part of the vocabulary of the language (and have seen, but forgotten countless words) chances are that even new unknown words somehow strike a chord in you, and then it will be much easier to remember them. You can use both target language dictionaries and base language dictionaries, - or best: do both types and find out what functions best for you.


As mentioned above repetition is an indispensable part of the process, and it should be done the next day (preferably) or just later on the same day. The repetition can of course be done in several ways, but in the two layouts above there are special columns for this purpose, - it is easier to keep track of your repetitions when they are on the same sheets as the original wordlists. However these column are only subdivided in two parts, one for the words in the base language, the other for the target language words. So you copy 5-7 base language words from the original wordlist, cover the source area and try to remember the original target language words. If you can't then feel free to peek, but - as usual - don't write anything before you can write all 5-7 words in one go. An example with Latin and English words:


Addendum ccccccc

The combined layout was the one I developed when I had used three-column wordlists for a year or so and found out that I had a tendency to postpone the revision - having it on the same sheet as the original list would show me exactly how far I had done the revision, and I would only have to rummage around with one sheet. And for wordlists based on dictionaries or premade wordlists (for instance from grammars) it is still the best layout. But I have since come to the conclusion that it isn't the most logical way to do the revision for wordlists based on texts, especially those which I had studied intensively and maybe even copied by hand. Here the smart way to work is to go back to the original text (or the copy) and read it slowly and attentively while asking myself if I know and really understood each and every word. I had put a number of words on a wordlist because I didn't know them so if I now could understand them without problems in the context then I would clearly have learnt something - and I would also get the satisfaction of being able to read at least one text freely in the target language. If a certain word still didn't appear as crystal clear to me then it would just have to go into my next wordlist for that language. So now I have dropped the repetition columns for text based wordlists.

Then what about later repetitions? After all, flash cards, anki and goldlists all operate with later repetitions. Personally I believe more in doing a proper job in the first round (where there actually are several 'micro-repetitions' involved), but it may still be worth once in a while to peruse an old wordlist. My advice here is: write the foreign words down, but only with translation if you feel that a certain word isn't absolutely well-known - which will happen with time no matter which technique you have used. The format doesn't matter, but writing is better than just reading - and paradoxically it will also feel more relaxed because you don't have to concentrate as hard when you have something concrete like pencil and paper to work with.

Memorization techniques and annotations[]

When you write the words in a word list you shouldn't aim for completeness. If a word has many meanings then you may choose 1 or 2 among them, but filling up the base language column with all sorts of special meanings is not only unaesthetic, but it will also hinder your memorization. Learn the core meaning(s), then the rest are usually derived from it and you can deal with them later. Any technique that you would use to remember one word is of course valid: if you have a 'funny association' then OK (but take care that you don't spend all your time inventing such associations), images are also OK and associations to other words in the same or other languages are OK. The essential thing in the kind of wordlist I propose here is not how you do the actual memorization, but that you are forced to do it several times in a row because of the use of groups, and that you train the recall mechanism both ways.

It will sometimes be a good idea to include simple morphological or syntactical indications. For instance English preposition with verbs, because you cannot predict them. Such combinations therefore should be learnt as unities. For the same reason I personally always learn Russian verbs in pairs, i.e. an imperfective and the corresponding perfective verb(s) together. With strong verbs in Germanic languages you can indicate the past tense vowel (strong verbs change this), and likewise you can indicate what the aorist of Modern Greek verbs look like - mostly one consonant is enough. There is one little trick you should notice: if you take a case like gender in German, then you have to learn it with each noun because the rules are complicated and there are too many exceptions. However most nouns are masculine, so it is enough to mark the gender at those that are feminine or neuter, preferably with a graphical sign (as usual Venus for femininum, and I use a circle with an X over to mark the neutrum). This is a general rule: don't mark things that are obvious.

Arguments against using word lists[]

Finally: which are the arguments against the methodical use of word lists in vocabulary learning?

One argument has been that languages are essentially idiomatic, and that learning single words therefore is worthless if not downright detrimental. There is a number of very common words in any language where word lists aren't the best method because they have too many grammatical and idiomatic quirks, - however you will meet these words so often that you will learn them even without the help of word lists. On the other hand most words have a welldefined semantic core use (or a limited number of well defined meanings), and for these words the word list method is a fast and reliable way to learn the basics.

Another argument is that some people need a context to remember words. For these people the solution is to use word lists based on words culled from the books they read.

A third argument is that the use of translations should be avoided at any costs because you should avoid coming in the situation that you formulate all your thoughts in your native language and then translate them into the target language. But this argument is erroneous: the more words you know the smaller the risk that your attempts to think and talk in the target language fail so that you are forced to think in your native language.

A fourth argument: word lists is a method based entirely on written materials, and many people need to hear words to remember them. This problem is more difficult to solve, - you could in principle have lists where the target words were given entirely as sounds (or as sounds with undertexts), but you would have serious problems finding such lists or making them yourself. But listening to isolated spoken words is in itself a dubious procedure because you hear an artificial pronunciation and not the one used in ordinary speech. However the same argument could be raised against any other use of written sources, except maybe listening-reading techniques.

A fifth argument: there is a motivational problem insofar that many people prefer learning languages in a social context, and working with word lists is normally a solitary occupation. It might be possible to invent a game between several persons based upon word lists, but it would not be more attractive or effective than the forced dialogs and drills used in normal language teaching.

Finally an example based on Dutch-Danish and Spanish-Danish (based on an older layout without repetition columns):



Of course there are alternatives to wordlists: the most extreme is the exclusive use of graded texts as the most vehement adherents of the natural method propose. I don't understand their motives, but respect their bravery. However I do understand the unorganized use of dictionaries plus genuine texts, but frankly I think there is room for improvement in that method.

Finally, there are well-structured alternatives like paperbased flashcards and electronic versions of these, all based on the notion of 'spaced repetition': Anki, Supermemo. However I can't give advice concerning these systems because I haven't tried them myself.

Main source for this article at Iversen's Guide to Learning Languages (HTLAL forum)