The Wikipedia Method

Here is how I do the bulk of my reading and translation as part of my process to learn a foreign language:  I use the Wikipedia method.

My Wikipedia method works this way:  First, I will go to the Wikipedia page of the language that I want to study.  For the Polish language, I go to Wikipedia Polski, and pick an article in a subject that interests me.  Of course, I open it in Google Chrome.  Then I open another tab with the same article, and Chrome will ask me if I want to translate it from Polish to English, and I will translate it.  Then I open another tab with Google Translate, set to translate from Polish to English, so I can translate individual words if needed.  Note also that I have added a link to a random article on Polish Wikipedia at the bottom of my blog.

I'll read through the article, and if I come to word(s) that I don't know, then I'll look at the version translated from the target language into English (this is better than looking at the same article in English Wikipedia, because usually that article is completely different, and the translated version is somewhat word-for-word).  Since the translation is sometimes a little off-kilter, if what I see doesn't make sense to me, then I'll go to Google Translate and enter the words one by one.  I have yet to encounter something that is so idiomatic and peculiar that I can't figure out what it means.

It's not as useful to translate from the source language to the target language (for example, if your native language is English, don't use the Chrome browser to translate from English to Polish [or whatever the target language is]). The reason is that the translation will be weird and your native language skills will be better at understanding a strange translation in your native language than in the language you are trying to learn.  Also, you want to look at grammatical and syntactical constructions (as well as translations) that are correct, and even though humans make mistakes, the version written by the person will be more correct than the version that is translated.

But, really, my Wikipedia method could be used with any source material that is in the Polish language.  I use it with Wikipedia the most because then I can learn about cool and interesting stuff too.  News sites and blogs in Polish on subjects that interest you might be particularly suited to this type of endeavor.  The only things that are really important are: 1) that the source code on the page of the site you are translating is not so whacked-out that it won't allow for a translation, or 2) that the site is not heavily dependent on images that deliver text.

Vocabulary Whore

I fully realize that I could probably become conversant faster if I would devote more study to things other than Polish vocabulary.  But the problem is, I'm a vocabulary whore.

There is a lot of time that I should be studying grammar, or listening to audio, or reading. And I realize that.  I do devote some time to those resources, but not nearly enough.  The problem is that I just like learning new words.  After I've studied a bunch, I always find a hundred more words that I would like to know.  So I add them to my flash cards.

Actually, they are usually there in the flash card deck already, so I just unsuspend them.  Sometimes they aren't, and I will add them to the deck.  And then I go back to studying vocabulary compulsively.

Today, I studied about 90% Polish, and 10% other languages (Dutch, German, Catalan, Italian, Czech and advanced Spanish--was there anything else? I forget).

I do devote some time to reading and translation.  I'll talk about the bulk of the reading I do in the next post.

Claims Of Extremely Quick Language Learning, And Studies Of Extraordinary Polyglots

Sometimes I hear claims that one can learn a language and be fluent in it in a matter of weeks or just a few months.  Learning a language is an extremely difficult and complex task, even for those who have mastered several languages.  That having been said, there are probably some individuals who are able to create neural pathways that are optimal for learning and are able to pick up a language in a very short period of time.

There is the well-documented case of Daniel Tammet, who apparently learned Icelandic in a week well enough to speak it on a talk show in Iceland (sidebar: Tammet's case was researched by Simon Baron-Cohen, a cousin of Sacha Baron-Cohen, the actor who played "Borat" and "Bruno").  Tammet had apparently learned several languages in a short period of time.  There is also Christopher, who displayed linguistic savant properties as well and could speak numerous languages.  Many such individuals are otherwise afflicted with developmental differences (autism, or some form of brain damage), and have their linguistic prowess as a "splinter skill,"

There are also some polyglots whose talent does not arise as a function of a developmental difference and are considered relatively normal in other ways.  There is a culture of talented polyglots who have been profiled; many of them started learning languages at an early age, but it is not unknown for people to begin learning languages as adults.  There have been some studies of what it takes to become a distinguished polyglot, and it appears that part of it is dedication and work, and part of it may be attributable to brains that are wired a certain way.  But there don't appear to be a lot of answers to whether such brain wiring is inherently innate, or can be achieved through learning.  Here is an interesting series of interviews with polyglots in which they talk about language learning.

There is also the matter of what constitutes a polyglot.  Some who call themselves polyglots may be somewhat faking it by knowing "just enough to be dangerous."  One would probably have to go through a testing regimen to determine exactly how proficient they are in a given language.  There are probably a lot of people who have various levels of proficiency in several languages that might fall short of fluency.

Now, let's get back to learning strategies that claim high levels of proficiency in short periods of time.  There is a book that I had studied from called "Polish in 4 Weeks: An Intensive Course in Basic Polish" (one of several texts that I have studied from, there is also a "Part 2" book for more advanced learners that I am currently going through), but I personally took a lot longer than four weeks to go through it.

The theory is that there are 28 lessons, and if you do one a day, you will have it down in a month.  There may be learners out there that can do that, but I certainly am not one of them.  I would venture a guess that the average learner would not be able to meet that one-month time standard.  You have to consider that the average student who takes two years of a language in high school or college can ordinarily barely speak the language, even if they made very high grades, unless they were extraordinarily motivated to actually learn the language rather than learn for the tests, which is possibly a different skill.  This isn't to say that the subject matter of the book is not good; actually of the numerous textbooks I have on Polish, I have to say that this is one of the better ones.

Enough for now, but I'd like to revisit this topic again at a later time.

Mixing It Up In Polish, Part 2

This post has been consolidated and moved to here.  Apologies.

Mixing It Up In Polish

This post has been consolidated and moved to here.  Apologies.

Polish Tongue Twisters

It seems like almost everything in the Polish language is a tongue twister to me.  But here is an awesome list of Polish Tongue Twisters.  I know I have trouble pronouncing the words "chrząszcz" and "źdźbło".  And then when you string two or three words together, it puts a whole 'nother dimension on some of these impossible pronunciations (not to mention full sentences and paragraphs).

Drop-Outs And Leeches In Spaced Repetition Learning

Drop-outs are words that suddenly seem to disappear from your memory.  This often happens when you are regularly adding many words to your lexicon on a daily basis.  You will have words that you are certain that you learned, but suddenly when you review them, you have no memory of having ever seen them.  It's like they got squeezed out of your memory by other words, or by other elements of your memory.

Sometimes a word will not be a total drop-out.  You will have some memory of some facet of it, but not retain an important relational structure that you had learned.  For instance, I talked about the situation earlier whereby sometimes I will have a word just suddenly pop clearly into my head, but completely disassociated with any meaning.  Maybe you will remember a word in this fashion, isolated from its meaning, when you had learned it pretty solidly and retrieved it multiple independent times in the past  Or maybe you can only make a connection to a meaning in one direction; for example, from the target language to the source language but not in the other direction (after you thought you had solidly learned it).  In this case, it will only be a partial drop-out.

Leeches are a different matter.  They are the black holes of learning, the words you just can't seem to get embedded in your mind no matter what.  There are several reasons a word may be a leech.  One of the most widely discussed theories is that the material you are studying is poorly designed for that word.  It may ask for too many associations.  For example, sometimes a word has more than one meaning.  Sometimes it has multiple discrete meanings, and those meanings are not grouped in ways that are common associations in your native language.  It may be difficult for you to grasp the meaning.  On occasion when this has been the case, I will make separate flash cards for the primary meaning(s) of the word and the secondary meaning(s) of the word, and then possibly consolidate them after I have learned both cards fairly well (of course, they can always become drop-outs at some point, and then become difficult again).

There is an excellent discussion of how to handle leeches on the Supermemo site, along with strategies for dealing with them.  For those of you who are not familiar with Supermemo, it is one of the first spaced repetition flash card programs, created by Piotr Wozniak, and there is a lot of discussion about how memory works on there.  You can really get lost there looking over the incredible repository of material on memory.

I subscribe to a sort of zen model.  For every component of learning that you need to do something for, there is something that you need to just as strongly NOT do to accomplish the task.  For every period that you spend learning, you need to have periods of rest and/or sleep, and these periods are just as important as the study.  In my model, leeches are words that absorb the negative energy from the other words so that those words can be learned.

There is also more discussion of what to do with leeches on the Anki site.  My thoughts on leeches are that when you are first learning, it is best to suspend leeches, as your learning will become more efficient without them.  Since a small amount of material will occupy a large portion of your time, it is best to move on to learning that will be easier.  Also, once you have moved away from whatever neural pattern you were experiencing due to the passage of time (another example of how NOT doing can enhance your learning), new patterns can arise that allow for easier learning.

However, once you get to the point where you have learned a lot of material and you are getting to more esoteric words, and there is more interference from other material after you have learned a large body of material, it is more difficult to drop leeches/  This is because almost EVERYTHING you will be learning can fall somewhat into the "leech" category, due to the fact that all the easier material selects itself out (because you have solidly learned it), and the fact that you will be working with more difficult concepts at this point.  This phase usually arises after having learned several thousand words.  At this point, you just have to accept that your learning will be harder.  But in some respects, it can also be easier in some ways too, because you will have a greater understanding of roots and syntax in the target language.

Both drop-outs and leeches are probably a function of retrieval, not of storage.  The material is in there somewhere, but the retrieval path has either not yet been completely constructed, or has been disrupted.

What Is Fluency?

The best definition of fluency that I have seen is, "the ability to read, speak or write easily, smoothly, and expressively." So you have become fluent when you can pretty much understand what people are saying and offer cogent and quick responses to what they have said to you, and do this back and forth without unnatural pauses.

That isn't to say that it is presumed that you can follow along with, say, a graduate level seminar in quantum physics.  The simplest threshhold of fluency could involve speaking at about a second-grade level.  You could still have large holes in your vocabulary that you are adept at working around; it's just that you can understand and respond without it seeming like you are struggling.

As an aside, though, I think of a grade level as more of an allegory than a precise measuring tool.  An adult will usually not speak on a second-grade level, strictly speaking.  Second-graders, can, for example, describe a lot of household items and know verbs that an adult might hardly ever use, like "to cluck" (gdakać).  How many times as an adult are you going to be called upon to say, "The cow moos?"  But a second-grader could tell you that easily.  Still, since it's not all that useful for you to know how to say that, learning to say things like this should be low on your list of priorities.  And, as an adult, you might know the word for "sociology" (sociologia) or "empathy" (empatia) relatively early on in your studies, especially if the related word is a cognate to the more complex concept that a second-grader might not necessarily know.

So there exists what I call basic fluency on the low end, and then native fluency on the other end.  Native fluency is what you would expect your average person who grew up in the country that speaks the target language to have.  And it can be uneducated native fluency at the low end and highly educated native fluency at the high end.  Most foreign speakers don't get to highly educated native fluency, ever. There are individuals who do, either because they are particularly gifted in language acquisition, or because they work like crazy to eliminate all the little errors in grammar and syntax, and deficiencies in their accents.  Many people who learn another language often have little grammatical errors or stilted modes of speaking in which sentences are constructed like they might be in a different language.  But note that grammatical (or spelling) errors occur even with native fluency.  Some are a part of colloqualism, and some run with a dialect.  And some are just common mistakes that native users of a language make.  Ain't a lotta English natives cain't figger out whut this sentence mean.  You see spelling errors in English all the time whereby some people use the wrong form of [it's or its], or [they're, their or there].  Other languages have little traps like that too.

Fluency doesn't just take one path; native fluency is just the most ingrained path.  Here is where it branches out.  You can have contextual (usually occupational) fluency.  This, at the low end, is where you might be able to communicate with a co-worker or an owner might be able to communicate with a hired hand using terms that are commonly used associated with the work (usually this type of communication is more faux-fluent that "real" fluent).  At the high end, it could be the specialized vocabulary that is used in, say, the medical or legal fields, or that a geologist or orchestra member might use.

I am fluent in Spanish.  I can communicate easily and freely in that language.  I would come somewhat short of saying, though, that I have native fluency.  I have a lot more than basic fluency.  I can communicate easily in speech or on the phone (speaking on the telephone is often a step up for language learners because they don't have the visual cues any more), with hardly any linguistic work-arounds.  I can understand almost everything I hear on TV, though sometimes I have to concentrate.  I can understand most words in the newspaper, though I would probably have to look up words much more often that I would have to in English.  So I still have room for improvement (but then again, I do even in English).  Have any of us really completely learned any language, even our own native language(s)?

I wrote a follow-up post on fluency here.

Phases Of Learning Words, And The Three Steps In The Illness Model

The last post talked about the skills involved in learning words and ways to develop those skills.  Now I want to address the phases of learning words.  All these terms are my own and I have no idea whether there are any similar concepts in the field of linguistics.

The first step is what I call simple recognition.  This is where you could pick out the word on a multiple-choice test among a few other choices.

The next step is isolated recognition.  This is where you can pick out the word on its own and associate it with the meaning.

Then you have reciprocal recognition, whereby you can pick out the word on its own and associate it with the meaning in both directions.

Up to this point, you probably still can't use the word on its own because you have only committed it to memory and not really worked up the retrieval paths in your memory.  Retrieval is one of the most difficult tasks.  You may find that words you think you absolutely knew just can't be called up when you need them.

I use an illness model of learning.  This point is what I call the "stroke victim" level.  You are trying to talk to people, and you know you have used the words, but you struggle to produce them.  Maybe nothing at all comes out.  Communication is frustrating.  The words are in there, somewhere, locked away in your brain, and you don't know how to get to them.  You may resort to pantomime rather than try to use words.  You might be able to muddle through at a basic level using every tool you can think of--writing it down, coming up for a word you know for it in another language, slowing down your speech or getting the person you are communicating with to speak more slowly, etc.  The one people try a lot (but is probably completely useless) is speaking louder and closer in their own language.  This is where you get the slightly annoyed blank stare in response, which is probably universal in meaning, but not where you wanted to go.

To break out of this level, you are going to have to get beyond the recognition phases and get to the recall phases.  And try not to drool.

Simple recall is where you can come up with the word when you need to most of the time.  This is a good thing.  You are getting to the basics of verbal communication.  It will become more fruitful to try to make yourself understood.

So your goal at this point is to graduate from "stroke victim" to "developmentally disabled".  This is where you can communicate, but you sound slow and stupid.  Your speech is halting and tentative, but a little more complete, and you can communicate basic needs and simple commands.  Me need bathroom.  Something smell bad.  Congratulations, you're now an idiot.

The next phase I call complex recall.  You can start placing the word in different contexts.  If it is a verb, you know many of the tenses and their forms.  If it is a noun or adjective, you know how to decline it.  You know how to use prepositions, but you are not perfect.  You are starting to grasp grammatical rules.

The next level in my illness model is what I call "backwards Alzheimer's".  This stage is long and drawn out.  This is where you have good days and bad days.  On the good days, you can understand and speak fairly well.  On the bad days, it's just not happening, and it gets frustrating.  And it's backwards, because your cognitive abilities in the language don't deteriorate more and more, but get better and better with more practice, until you have good days more and more often. Rather than experiencing "the long goodbye", you are in the midst of "the excruciatingly long hello".  Unless you really have a degenerative brain disorder, in which case you probably should get a medical opinion.  Sometimes I wonder.

When you are having mostly good days, you will probably find that for individual words, you have what I call "approaching word fluency".  Before, you were basically learning to count cards at the blackjack table.  Sure, you know how it's done, and you get all the concepts, but it takes practice, practice, and more practice to get it down so you can do it smoothly without any tells so you don't get your legs broken by the pit boss.  You need to just keep working it, and use your mind like a wrench on that recalcitrant bolt that constitutes the word you are grappling with.  Keep doing this until the rusty equipment is dismantled and put back together over and over again.  This is where the elbow grease effect kicks in.

Then you have what I call "word fluency."  This is where you can really work it easily for this particular word.  When you have this with about four thousand words, and you can interconnect them without pauses, you start approaching language fluency, which I want to address later.

Skills Involved in Learning Words

There are a lot of different skills that come into play in learning words.  First, there is learning the word itself.  That is, just placing the sounds of the word in your head.  I can't tell you how many times a random word in a foreign language has just popped into my head suddenly, without any associated meaning.  The word itself is just simple mellifluous cacophony superimposed on my conscious processes, floating without an anchor to any frame of reference.

Then there is the association with a meaning.  Sometimes the memes associated with the words in each language don't mesh up on a one-to-one basis, and sometimes the meaning is slightly off from one language to the other, or there is really no exact correlate in another language.  For example, the word "chciwy" means, literally translated, "want-y" (chęć is "to want").  But the closest translation is "greedy."  Or the word "gwałtowny" means "rape-y" (gwałt means "rape").  But it is translated as violent, torrential and sudden as in a sudden storm.

So you have to work up the association with a meaning between the target language and the source language, and then do it also in the other direction.  You just need to keep working it and working it in your head, and find as many tricks as you can to implant the associations in both directions.

One thing that is often recommended is to try to think of some sort of mnemonic or a word picture that helps you learn it, and the more silly, ridiculous or vulgar the mnemonic can be, the easier you will remember.  For example, I learned the notes of the lines on the bass clef line many years ago with the mnemonic, "Good boys don't fuck animals."  I dare you to get that out of your head.  I read an interview with a memory champion once where he said that some of the images he used to remember items were just really, seriously wrong.  He mentioned things like deviant acts that his grandmother was involved in.

OK, maybe that doesn't work for you.  One trick I use a lot is to look at the patterns of vowels in each word and compare them in as many ways as I can.  Look at the ones they have in common, look at how many vowels are in each word, think about their order, etc.  If that doesn't stick, I'll do the same with the consonants, though that is usually harder.  Sometimes I'll take a walk while I memorize words and try to recall what I was looking at when I thought about that word (sort of a soft version of the "method of loci", which I never really have much taken to in its pure form.

Verbal repetition works for me in some cases.  Often if I take ten minutes to just repeat the English word and the Polish meaning over and over again, working it around in my head, I will know it pretty well.

Sometimes I will be thinking about a Polish word and all of a sudden a root part of the word that is similar to another word that has that root pops into my head as an epiphany.  That almost always helps with learning the word.

It helps to approach the word as a thing of beauty or a piece of art to be appreciated, rather than an obstacle that needs to be overtaken.  Polish is beautiful and intricate in the manner in which its structure is put together.  Polish grammar is like high performance art mixed with some really fucked-up and arbitrary (pseudo) science.  It works better for me to admire rather than memorize.  The memorization will come, but it is better if it comes as an organic, natural process, instead of the pounding of pegs into holes.

Sometimes I encounter difficult words that have lots of lexemes (lots of disparate meanings).  If this is the case, sometimes I will study them in a separate group, because they require a lot of mind-massaging to be remembered fully.  And it is this class of words that reminds me the most that the most difficult part is creating the retrieval paths in the brain.  That stuff all goes IN easily.  It's in there somewhere, rattling around in the brain.  It's finding it and being able to spit it out that requires the most effort.  One proof of that is that when one is learning flashcards, one will learn the meanings of the word pretty easily at first, and be able to retrieve them for the first few days.  But after that flash card has been retired for a while, and one comes back to it, one finds considerable difficulty, and has to recreate some of the pathway of retrieval.  It's usually easy, then hard (but not as hard as originally), then easy, then hard (but less difficult)...it evens itself out, becoming less difficult in each iteration, until the word is learned, that is, until the retrieval occurs in a consistent fashion.

One visualization technique I use when I'm consistently not able to remember some of the lexemes of a word, is picture the neural connections between where the word is stored and the output of the lexeme that I am trying to retrieve from it.  I picture the place in the brain where the word lies, the nerves with their dentrites and axons as conduit from the place of origination to the destination, and trying to get that lexeme from the dusty, dank place in the brain where is is stored, to the proper output vehicle (I picture the mouth, because one would be speaking the result), utilizing the fanciful neural path that I am imagining.

Maybe I'll revisit this topic later.  I also want to talk later about "leeches" (words that are particularly difficult to memorize) and "push-outs" (words you really thought you had learned but have suddenly disappeared or become difficult). I want to talk about "practicing for diversion", which is a technique I use whereby I will work on learning some new words not necessarily to learn them, but as a diversionary tactic for my brain so I can go back to my original list after wiping my memory to see how well I really learned the words I was trying to learn.

A Primer On My Polish Studies

I've decided to start a blog about my Polish Language studies.  I've seen some other blogs out there of this sort and they have been helpful to me, so maybe this will be helpful to others as well.  I've made a few comments about the Polish language on David Morrison's most excellent blog about moving to Poland, No Home But The World, and maybe I'll incorporate and expand on the thoughts I've put there eventually.

I've been studying Polish for a little over two years now.  It's an incredibly difficult and complex language, and I would liken learning it to maybe doing a hundred-thousand piece jigsaw puzzle.  I do have some experience with jigsaw puzzles, because a while back, a large group of people including myself completed a jigsaw puzzle with over ten thousand pieces, and it took us all over a year.  Of course, that was just attacking it sporadically, a few minutes a day.

I've been trying to study about two hours a day on average, which is a pretty large amount when you average out the fact that sometimes I will really bear down and study six hours a day, and sometimes I will burn out and not study for a couple of weeks.  The best and most productive thing I have found to do is to study for about four or five half-hour increments a day that are spaced out from each other.  On flash cards alone, I have averaged 735 cards a day over the life of my flash card deck, but only 470 a day over the last year (I know the exact number because my flash card program spits out all kinds of statistics).  And that's just on Polish...I'm also studying Dutch, German, Italian and Danish, but not with near the ferocity that I am applying to my Polish studies.  I probably spend about ten hours of Polish study for each hour of all the other languages combined.  Still, I had read on another Polish learner's blog that he studies about six hours of Polish for each hour of German and still knows German better because Polish is so complex and arcane.  I can relate.

And I strongly recommend Anki as a flash card program.  It's free and open-source (though donations are strongly encouraged and certainly deserved due to the excellent quality of the program), and there are a huge number of shared decks created by others.  But beware, the decks can vary in quality (since anybody can make them, they might not follow the best principles for learning; e.g., they might include too many facts on a card or just have cards for one direction of translation).  The Polish deck that I am using is excellent.  It was created by Per Ericksson, who really created one of the best language learning decks I have seen.  And I have added my own material to it so my deck is about 150% of the size of his (I have nearly 15000 facts and 30000 cards in mine; there are two cards for each fact because there is one for Polish-English, and one for English-Polish).  I emailed him many changes and corrections I had for a while and he incorporated them, but I guess he has not been maintaining it as diligently lately.  I don't want to put mine in the shared area because I don't want to supercede his, but if you want a copy of mine, let me know in a comment and I can send it to you (actually, I can send you a link to it so you can download it because it is so large it won't send in an email under most circumstances).  Another great thing about this deck is that I use it as a dictionary of first resort as well, by using the search function.

So after two years of study, I am probably at a low B1 level in reading and a mid-A2 level in speaking, and maybe a low A2 in understanding spoken speech (using the Common European Framework of Reference for Languages).  My speech skills (especially in understanding rapidly spoken speech) lag far behind my reading skills as I haven't been exposed to a lot of spoken language, and as my study methods are probably emphasizing vocabulary at the expense of grammar and verbal skills a lot more than a language course would make me do, mostly because I'm obsessed with knowing words and two- and three-word combinations of words in short phrases or idiomatic constructs.  So I'll probably learn about ten thousand words and idioms pretty well, then slowly start moving to more correctly using them and understanding them in speech.  That is probably not the way most people would learn...most would probably become "fluent" in a hundred words, then move up to a thousand, etc., but learning in a more balanced fashion and emphasizing verbal usage more.  Everybody finds their own path.

Anyway, that is probably enough for now.  I hope to maintain this blog at least sporadically; knowing my history of blogging there may be times when I am idle for a while.  But if somebody bugs me to put up a blog post, or suggests a topic for one, I might be spurred to action.