Tuesday, December 30, 2014

How to Make a Gold Duolingo Tree

Finishing Duolingo isn't enough; you want all the skills to be "gold," and you want them to stay gold. In this post, I'll talk about how to "regild" a completed tree, what to do about weak skills if you're half-way through a tree, and how to keep a tree gold from the very start. I'll also discuss how to do this in a way that actually teaches you something--it shouldn't be an exercise in aesthetics!

Those who are impatient can just skip straight to the end and read what I recommend as the best practice. For everyone else, I'll explain what it is we're talking about and how I reached the conclusions I did.

The Decaying Tree

When you first finish a skill, Duolingo turns its icon gold, and five "strength bars" (like on your phone) appear beside it. If you don't do regular "strengthening" exercises, though, the strength bars will gradually weaken from five to one, and at anything less than five, the skill appears in its original color--not gold. This is what we mean by "strong skills" and "weak skills." "Regilding the tree" means doing enough exercises to turn the whole tree gold again.

People who do new lessons but never do review lessons get used to the way a wave of color chases them down their tree. Those who took the placement test and skipped over a large part of the tree are often stunned when a huge slab abruptly turns color all at once. In both cases, Duolingo is sending a message that those are skills that ought to be reviewed. This post is for those who want to do something about weak skills and who want to enhance their language learning at the same time.

Why Bother?

As Duolingo warns, if you don't review what you learned, you'll start to forget it. Lots of people eventually abandon Duolingo because they get about a third of the way down the tree and then they have terrible trouble finishing new lessons, due to the fact that they've forgotten too much of the earlier material. Doing regular review lessons spares you from this.

Duolingo has a decent algorithm that tries to target the things you most need to review. Done right, the review lessons really shouldn't end up being a waste of time--they should help you solidify your understanding of the language. And they often contain material you didn't see when you did the skill for the first time.

Two Kind of Strengthening Exercises

To reverse this decay process, you can either do a skill-specific strengthening exercise or a general strengthening exercise. To do a skill-specific one, you pick a weakened skill and click on its icon. On the right, you'll see a colored box showing the strength of the skill, and inside that box is a button to strengthen the skill. Press that button and Duolingo gives you a review lesson targeting the skill as a whole. That is, it doesn't just repeat one of the lessons--it asks you questions drawn from all of them.

Because it only takes 17 correct answers to complete a lesson, and because Duolingo usually has far more than 17 possible questions per lesson, when you do a review lesson you will frequently see sentences you never saw when you did the skill in the first place. Usually not entirely new words, but you'll definitely see new ways to use the old words.

Depending on how weak the skill is and how big it is (i.e. how many words it covers), a single strengthening exercise may not be enough to make it gold again. Repetition may be necessary. On the other hand, occasionally you'll strengthen some entirely different skill that just happens to share a few words with the one you were trying to strengthen.

To do a general strengthening exercise, you go back to the main menu and click the blue "Strengthen Skills" button on the far right of the screen." This brings up a review lesson that has a random mix of things you have already studied but that Duolingo thinks are weakest.

Because the benefits of the general exercise are spread across the entire tree, it usually strengthens anywhere from zero to three skills--some people have reported as many as 13.

In short, your tree weakens over time, and you have to do strengthening exercises if you want it to stay (or become) gold. The question is how often to do them and what kind to do. To answer that, we'll start with a look at Duolingo's algorithm--what's known of it, anyway.

How it Works

Duolingo does not publish much about the algorithms they use for weakening and strengthening, but they did make at least one posting with hints about it. Further, a Duolingo user with the handle pinkodoug has made an extensive study of the Duolingo JavaScript and he has some plausible ideas about their algorithms for strengthening and weakening.

Drawing from those sources plus my own experiments, here's my best guess as to what they do. (N.B. these are my conclusions, so if they're wrong blame me--not the sources above.)

To Duolingo, strength is about words, not skills. Strong words have a score of 100%, and they decay over time until they reach zero. All words are decaying all the time, but some words decay faster than others.

You can actually see this if you click on the "Words" tab at the top of the home screen.

This shows all the words that Duolingo thinks you've studied so far. It tells you how long ago you practiced each one, and it gives each word's current strength using a bar system that goes from 1-4 (not 1-5 the way skills do). If you sort by "Last Practiced," you'll see that the older words tend to be weak while the latest words are almost all 4-bars strong. This is because all words weaken over time, but, as I said before, some words decay faster than others. We'll get back to this last point in a minute.

Every skill is associated with a set of words--probably the same words that you see in the individual lesson descriptions. Whenever those words have an average strength of 2.5 bars or more, the skill is gold. Otherwise, it shows the original color, and the skill-strength declines based on some function of the average word strength.

Whenever you do a review lesson, it targets some number of your weakest words for review. A skill-specific review only picks words from that one skill. A general review picks words from every lesson you've completed so far. Otherwise they're the same--with one important exception: in the tips that Duolingo offered, they said that skill-specific strengthening turns your tree gold faster than general strengthening does.

As long as you complete the review lesson, the targeted words all get set back to four bars whether you get them right or not. The catch is that getting them right makes them decay slower in the future. Getting them wrong or using the hints makes them decay faster. This is why some people report making tremendous efforts to strengthen every single skill only to watch them start decaying again a few days later.

The fact that some people seem to keep their trees gold without much effort suggests that there actually is an optimal way to do this. To determine what that might be, I did a few experiments--with help from a friend.


Greg's French Tree

For the past twelve months, I have been slowly going through Duolingo's French tree, following a rule of doing two review lessons per day followed by one new lesson. In that time, I have completed most of the tree, and it has never weakened. That is, although an individual skill has weakened two or three times, I simply strengthened it specifically and thus ended the day with a solid-gold tree. During that time, I had three different week-long vacations during which I couldn't do lessons. Even so, when I got back, the tree was still gold. Others have reported similar experiences.

This suggested to me that it is possible to have a durable gold tree without a whole lot of work. That is, it seemed that as few at two general strengthening exercises per day might be enough, if done daily for a long enough time.

Naomi's Italian Trees

My friend Naomi in New Zealand was interested in exploring this, and she agreed to help do a few experiments. She had finished her English-to-Italian tree (aka the "forward tree") in just a few months and wanted to gild it. She had also just completed the Italian-to-English tree (aka the "reverse tree") in a single day and wanted to keep it gold.

The experiment was very simple: On her forward tree, she did two general strengthens per day. On the reverse tree, she did three general strengthens per day. Back when Duolingo allowed you to fail, she would persist until she completed that number, although, as it happens, she rarely failed an exercise.

Here's what happened.

With three general strengthens per day, the reverse tree got worse for about 40 days, leveled off for about a month, and then started improving in an irregular sort of way.

The forward tree decayed steadily, reaching 57 weak skills on day 91. At that point, we changed our strategy.

Starting on day 92, Naomi began doing three skill-specific exercises instead of two or three general ones. She selected the skills to strengthen by picking the weak skills highest in her tree. She applied the same algorithm to both trees.

We only have 13 days of data with the new method, but the results speak for themselves.

This graph shows how many weak skills were left each day after Naomi did her strengthening exercises. This shows that the reverse tree was 100% regilded on day 99 and that it has remained so for six days now. On the present trend, the forward tree will be regilded in two more weeks.

Obviously this strongly suggests doing the per-skill exercises rather than the general ones--just as Duolingo had advised people. In fact, the per-skill exercises appear to be more than three-times as effective at regilding. It also suggests that two exercises per day was not enough.

We adopted one additional rule: substitute general exercises for specific ones if there are no weak skills left, but never do more than two general exercises on the same day. That is, if the tree has any weak skills, then Naomi still does three strengthening exercises--padding them out with general ones as needed, but if the tree was solid gold to begin with, then she only does two generals, not three. The idea is that, like with my French tree, a durable gold tree should eventually need only two per day to maintain it.

Note: Throughout the process, Naomi reported that the algorithm seemed to have a knack for finding exactly the things she was weakest on. She felt the exercise improved her Italian enormously, even if she did spend 91 days making no visible progress on regilding her trees.

Greg's Spanish and Italian Trees

During this time, I let my own Spanish and Italian trees decay. (Both had been gold on day zero of Naomi's experiment.) Over the next month or two I'll try to reproduce the results from Naomi's trees, but I wanted to share one key result that I got on the first day:

For the best learning experience, you should strengthen from bottom-up instead of top-down.

The reason to prefer bottom up is that it drills you in the skills you most-recently learned--the skills you are most likely to need help in. You'll discover that Duolingo has a lot more material for those skills that you never saw when you did them the first time. Most people have a weak grip on the last skills in a tree, so it makes a great deal of sense to prioritize strengthening them first. In contrast, the stuff at the top of the tree is far easier to do, but it's also far less valuable to review.

The exception would be for someone trying to resurrect a long-unused tree or for someone who has become overwhelmed and is considering quitting. Those folks should strengthen from the top down because it's equivalent to starting over, in a way, without losing everything you learned the first time. (In my view, no one should ever delete their Duolingo tree and start over.)

Other Issues

I always advise people to avoid timed practices and not to use the mouse-over hints unless absolutely necessary. I'll explain my reasoning for both suggestions.

Timed Practices

You buy the ability to do timed practices at the lingot store. It's a one-time purchase, but it's still a bad deal. The way it works is that when do a review Duolingo offers you the option to do timed practice instead of the normal kind. In fact, it becomes the default.

In timed practice, you only have a few seconds before the clock runs out and you lose the round. Every correct answer puts more time on the clock, so you need to go like lightening to keep ahead. In theory, this helps you develop the skills to handle conversation in real time.

Many of the people I've talked to who reported doing lots and lots of review lessons with little progress on regilding their trees told me they were doing timed practices. I think this strengthens the tree slowly because time pressure encourages people to skip long sentences. The earlier post from Duolingo said that every word counts, so it would make sense that skipping the long sentences would really hurt you.

Given that Duolingo's new UI is much more forgiving of mistakes, it probably makes more sense to try to get the same effect by simply going through the exercises as fast as you can, accepting the fact that you'll make more errors. As a bonus, when you do make an error, you can stop and study what you did wrong. Anyway, if you want to regild your tree, you should probably stay away from timed practice.


The other thing that seems to be strongly associated with lack of success is using hints. Using a hint means doing a lesson and mousing over a word to get Duolingo to show you the definition.

Unfortunately Duolingo takes this as a hint that you don't know that word at all, and it makes that word decay faster. Do use the hints if you really don't know the word. (We want to help Duolingo figure out what you really need to study, after all.) But don't use them just because you want to be 100% sure of a word before you press return.


Plan to do three exercises per day. If you can't invest that much effort, you probably can't do this in a reasonable time. With that, here is the system I recommend for making durable gold trees with Duolingo:
  • Always do a skill-specific strengthening exercise if there are any weak skills in your tree at all. Only do a general strengthening exercise if there are no more weak skills to do.
  • If there are multiple weak skills to choose from, always pick the one that is furthest down the tree. (That is, prefer skills you learned later over ones you learned earlier.) But, if you are effectively starting over (e.g. after a long absence or because you're overwhelmed) then pick the highest (that is easiest) weak skills first.
  • For an incomplete tree, do two strengthening exercises and then do one new exercise every day. If you're not ready for a new exercise, do a third strengthening exercise instead.
  • For a complete tree that isn't all gold, do three strengthening exercises every day, using general strengthening exercises if you run out of skill-specific ones.
  • For a complete, solid-gold tree, only do two strengthening exercises per day--both of which have to be general ones, of course.
Never do a timed practice. Never use hints.

Finally, take your time. Don't try to regild your tree in a day or two of inhumanly focused effort--followed by a month of neglect. A steady effort over a long period of time is the best way to do this. You'll learn the most that way, and it won't burn you out.

If you follow this plan, assuming that you are doing your best on the exercises, your tree will eventually be "durable gold," meaning it will stay gold with minimal effort on your part, even if you go on vacation for a week or so. How long it will take to get there depends on a lot of things, but a month or two seems typical. (It depends on how well you do the exercises and on how deeply decayed your tree was in the first place.) Have at it, and best of luck!

I would love to hear from anyone who gives this a try. Leave a comment and let me know how it went.

Sunday, December 14, 2014

Como Agua Para Chocolate

I finished Como Agua Para Chocolate ("Like Water for Chocolate") this week, and enjoyed it a lot. Other than requiring the reader to master a hundred or so words of cooking vocabulary, it should be well-suited to an intermediate reader with a good sense of humor. Young female readers may especially enjoy it, but it's really a fun read for anyone.

Outline of the story (no spoilers)

A basic but boring description

Tita la Garza is the youngest daughter of a wealthy Mexican family whose ranch lies in northern Mexico not far from the US border. The novel takes place during the Mexican Revolution in the early 1900s. Tita wants to marry Pedro Muzquiz, the boy next door, but her family's tradition says that the youngest daughter must stay home to take care of her mother. She rebels, but her mother is a very strong woman who brooks no disagreement. Tita boils with repressed anger and passion--like the water one boils to make hot chocolate.

A more exciting description

Strange things happen around Tita. Magical things. Surreal things. Forced to work in the kitchen, she pours herself into the work (and excels at it), but when she's very emotional--for better or worse--all kinds of things happen to people who eat her cooking. Even without Tita's help, the la Garza ranch is a place where people do strange things for stranger reasons, but Tita definitely pushes it over the top.

Strange things happen to the reader too. You laugh. You get angry. You get sad. And then you laugh some more. 

There are twelve chapters, named for each of the months of the year, although the story encompasses more time than that. Each chapter is organized around a particular Mexican dish that Tita will prepare, and you can bank on the fact that something unexpected will happen when people sit down at the table with it. 

The target audience

Unquestionably, this is a girl's book. The heroine is a teenage girl, all the strong characters in the book are women, and cooking is a major part of the story. That said, you don't have to be a girl to enjoy it, and there is a lot here to enjoy. For me, at least, the book flew by. I don't think I was ever bored.

Challenges Reading the Novel in Spanish


Como agua para chocolate has quite a lot of cooking vocabulary. You'll find the names of cooking utensils, actions for preparing and cooking food, and a long list of foods. Assuming you are reading on a Kindle, you'll find that the free monolingual dictionary almost always has definitions for these words, so be sure to check it whenever your bilingual dictionary lets you down.

A trick that is sometimes useful, especially for long lists of unknown words, is to use the translate feature on the Kindle. Although the translation often makes a hash out of complex sentences, when you have a simple list of nouns, it usually does a pretty good job, and it's far faster than looking up seven or eight words in a row.

There are a few references to the Mexican revolution that might throw you. One important one is that a follower of Pancho Villa is a *villista*, but none of my dictionaries told me that--I just had to figure it out.


For this novel, I tried an experiment: during the first chapter, I highlighted every word that I had to look up in the dictionary. I sometimes look up words even though I think I know them, so I didn't highlight anything if the result was what I expected. Just the words I really needed to understand the sentence.

At the end of the first chapter, I created a new deck of flashcards with Anki but instead of using my special template, I used the basic Anki template--the one that assumes you only drill in one direction and that you don't type the answers. On the front sides, I put the Spanish words I had looked up, and on the reverse I put the definitions.

Drilling on those words was really fast. For example, Anki might prompt me with:

I might think "that that means "to shell," and I'd press the space bar. Anki would show me the answer:

In this case, I'd be right, so I'd press space again to mark it correct and advance to the next card. If I was wrong, I'd press the "1" key, marking the card wrong for further review.

Although not as comprehensive as the sort of two-sided cards I create for serious vocabulary review, I could blitz through these one-sided cards like lightening, largely because I didn't have to type anything.

After drilling on the words from the first chapter, I proceeded to read the second chapter. Sure enough, lots of those words occurred again--even unusual ones. I repeated the experiment in chapters two and three. This fairly simple exercise made the rest of the book a good bit easier to read. I'll try this again with the next couple of books to see if I really think it's worth the trouble.

Amazon actually has a built-in flashcard feature called "Vocabulary Builder" which does almost the same thing. Unfortunately, as of this writing, there is a bug in the Vocabulary Builder that causes your Kindle to slow down and eventually lock up if you look up too many words with it. I can't get past a single chapter without having to turn it off.

Friday, December 05, 2014

Finding Foreign Novels to Read

One of the challenges of reading in a foreign language is finding novels to read in the first place. You want something hard enough to challenge you but not so difficult that you give up in despair. If you already have recommendations from friends or teachers, that may be all you need, but recommendations can go badly wrong too. (See the last section of this post for a story about that.)

Even if you follow my advice on reading a foreign novel with a Kindle, you are likely to spend a hundred hours or more on this book, especially if it is your first in the language. It's worth your while to take a couple of hours to pick one out.


Because my strategy calls for using a Kindle, it does limit you to books you can buy on a Kindle. This turns out to be a limitation, but not a fatal one.

Amazon has a large selection of foreign-language novels in the Kindle store, but you'll be disappointed to learn that it's only a fraction of what's sold in those markets. For example, I checked just now and there are 113,182 Spanish Kindle books in the US Kindle store, but on Amazon.es, there are 3,198,321 available. That calls for an explanation.

Unless you have a mailing address in the EU, Amazon cannot legally sell those books to you. This is because standard contracts with publishers set out different rules for different countries and those contracts usually only give Amazon the right to sell a book in a few markets. It's okay (in theory) to order copies of the physical, printed books, if you're willing to pay the shipping costs from Europe, but you cannot buy the e-books. The fault for this lies entirely with the publishers, who generally won't sign international contracts even for e-books. People complain about this a lot.

As a practical matter, there are plenty of interesting books in the US store; you really aren't going to run out. But sometimes I'll see a book review that praises the English translation of a foreign novel, and I'll want to see if I can read it in the original language. So far, I have never been able to.

The most you can do is write a letter to the publisher. Short of that, you can find the US Amazon page for the printed book and click the link that says "I want this book on the Kindle." 

The upshot is that you have to start your search with the set of books that are actually for sale on Amazon in the US.

Selection Criteria

As mentioned above, you want to pick a novel that'll be fun and challenging but not too difficult for you. Toward that end, I've come up with a set of rules. They aren't hard-and-fast rules, but they're a reasonable guide to start with. As you read more and more books, you can relax some of these rules.

The most important thing--above all else--is that you pick a book that interests you enough to stick with it. This is a principle that supersedes any rules.
  1. The book must be an "authentic novel." This means written by and for adult native speakers. That rules out translations, children's books, and materials for students. The reason is that these things all dumb-down the vocabulary and sometimes use unnatural grammar.
  2. The book must be new to you. If you already read an English translation of the book, don't try reading it again. It's not just that this makes it too easy--it also makes it too boring. Worse, because you already know what you're reading, it can lull you into a false sense of accomplishment--of thinking that you understand what you're reading when you really don't.
  3. The book should be exciting. I like crime novels, action-adventure, etc. Find something that sounds fun. If this is your first novel in the language, you'll need something interesting enough to get you through the first chapter. Young-adult novels may be a good compromise. They're only a little simpler than mainstream novels, and they're action-packed.
  4. The book should be contemporary. Books over 100 years old often use different grammar and/or spelling. Sadly, most of the free books on Amazon are old ones.
  5. The book should not be too hard. Literary writing uses a bigger vocabulary and often uses unusual grammar. Science Fiction often makes words up (some claim that this is offset by the fact that SF is otherwise simpler).
  6. The book shouldn't be too long. Under 200 pages is good. Over 500 is probably too much.
  7. The author should be one you haven't read before. Authors have favorite words and phrases that they reuse a lot. That's part of why the first chapter or two of a new novel is always the hardest to get through. But if you keep reading books by the same author, you're not growing. That said, if you really like an author, that can add a lot to your motivation to read.
  8. The book should be a professional publication. You do not want to be reading something with spelling or grammar errors in it! Almost all self-published books are really bad, and it doesn't matter what language they're written in.


Start by going to the foreign-language section of Amazon's US Kindle eBooks store. If you're in the UK, then visit the UK store.  Amazon Canada is more complex because French Books are listed separately from Foreign Books. In any case, look at the left-hand column and select the language you're interested in. That will then let you select the genre (e.g. crime stories, science fiction, romance, etc.) and finally you can look at the top twenty or so best sellers.

With that restriction, start at the top of the list and ctrl-click on every novel that seems like it might meet your criteria. This opens up a separate tab for each one. When you get to 10 or 15, go through them and discard any that don't meet the criteria above.

Now take some time and read the book descriptions on Amazon. Discard anything you're sure you wouldn't be interested in. (For example, I won't read anything that depends on some conspiracy theory being true.) Then have a look at the comments from readers--especially the three-star ones. To get more comments, go to Amazon's foreign-language site (e.g. amazon.es for Spanish) and read the reviews posted there. You can't buy anything from the foreign language site, but it's perfectly okay to read the reviews. Be suspicious of anything that doesn't have very many reviews. Goodreads is another place to look.

Of the books that remain, do Google searches for the titles and/or authors. See if they have Wikipedia articles that say anything positive about them. Look to see if they won prizes.

For any book that passed your screen, add it to a Wish List. Amazon lets you create private, named wish-lists to keep track of books you're thinking about but haven't decided to buy. Sleep on it. Review the list in the morning, and buy one!


Conventional Wisdom

Many people will tell you not to attempt a book if you need to look up more than five words per page. You'll get discouraged and give up. There is a lot of truth to this recommendation--it certainly describes my own pre-Kindle experiences. But I claim that advice is now out-of-date for the following reasons:
  1. A Kindle with electronic dictionary eliminates 99% of the work involved with dictionary lookups. That completely changes the equation.
  2. The notion of "a page" was always shaky (would that be a page from a hard-back book or from a paper-back book) but on a Kindle it makes no sense at all. Scholars studying reading usually talk in terms of words. To read a novel without a dictionary, you can only be confused about 1 or 2 words in 100. I would adapt this rule, then, and say that there should only be 1 or 2 hard words in every 100. That means words that take more than a few seconds to figure out even with the dictionary.
  3. If you use the old standard, there are almost no books that an intermediate student can read. The ones that are available are called graded readers, and they use a carefully restricted vocabulary. They're very expensive, they're rather boring (in my opinion), and there are very, very few to choose from. (Most of them are for people who are trying to learn English.)
  4. It makes more sense to estimate the difficulty after the first chapter. Unless it's really easy or the author is familiar to you, the first chapter will be by far the hardest. It's not good to give people reasons to give up so soon.

A Cautionary Tale

A friend or a teacher may recommend a book to you. You should welcome that, but check it out a bit before you read it. I'll illustrate just how bad it can be.

When I was a senior in high school, the teacher of my fourth-year Spanish class assigned each of us a novel to try to read over the Christmas holidays. She gave me La Gaviota by Fernán Caballero. For all the reasons I've described earlier, I failed to finish even the first chapter. Thirty-eight years later, I am still ashamed to admit that I lied to her about it, claiming I'd read it but that it was just really, really dull.

In the summer of 2014, after I'd mastered the technique of reading a foreign novel using a Kindle, I decided that after all these years I would finally make this right, so I bought a copy of La Gaviota and read it all the way through. Even with all the resources of modern technology, it took me weeks to finish it.  Part of the problem was that it was written in 1848 and the grammar and vocabulary were different enough to make quite a few sentences difficult to decipher. I eventually broke down and bought an English translation, which I consulted when all else failed. (The translation turned out to only cover the first half of the book, but fortunately I didn't need it after that.)

And it was awful. It's about a peasant girl whose talent for singing briefly lifts her to the top of Spanish society but whose selfishness and ingratitude destroys her life and the lives of those around her. There isn't a single likable character in the story. The book has an unfinished feel to it, at least by modern standards, because it has numerous plot elements that go nowhere and lots of characters who are introduced but never developed. Worst of all, the novel's message seemed to be that God was punishing her for rising above her place and punishing those her helped her do it. Yuk! (For a different opinion, read Eva's glowing review of La Gaviota.)

To add insult to injury, the book was actually a translation from French into Spanish, so it wasn't even an authentic Spanish novel in the first place.

On the bright side, I no longer feel quite so ashamed of myself. Mrs. Parker never should have assigned such a book to an 18-year-old boy. Not without strong guidance on how to read it, anyway.

So pick your novels carefully. If you like, you can review my own list of novels. These are books I've read, am reading, or thinking about reading.

Saturday, November 15, 2014

Sostiene Pereira (According to Pereira)

This month, I finished reading the literary novel Sostiene Pereira (Antonio Tabucchi, 1993, 214 pp). It's a suitable novel for an intermediate Italian student looking for a first novel to attempt. It's a quality piece of writing, the vocabulary isn't overwhelming, and the story is entertaining.

I attend a monthly Italian conversation meetup here in Seattle, and at our last meeting I had asked for suggestions for an Italian novel to read. Previously, I had read two Italian detective novels, but I thought I was ready for something a bit more serious, and Sostiene Pereira is what the organizer proposed.

A Short Review

Sostenere means to support, or to affirm, or to maintain, but in the context of the story, "according to Pereira" is probably the best way to translate it. The first sentence in the novel is therefore, "According to Pereira, he met him on a summer day." It quickly becomes clear that the third-person narrator is some government official writing a report to a superior. This style creates a certain tension because we expect Pereira to have serious problems, and we're not even sure if he'll survive the story.

Problems aren't hard to find in fascist Portugal in 1938. Pereira is the middle-aged editor of the Cultural Page of a newspaper in Lisbon. He meets a young writer who reminds him of the son he and his late wife were never able to have. But the young man writes things that are too revolutionary to publish, and he has a girlfriend who is illegally helping the Spaniards fight Franco. Against his better judgment, Pereira does what he can to help the couple, even though the receptionist in his office seems to be working for the secret police.

I thoroughly enjoyed it, and I believe it would make an excellent first novel for a strong student of Italian.

How I Went About Reading It--Dealing with the dictionary

I have already described my technique for reading a foreign novel on a Kindle, but Italian presents some unique challenges--mostly involving the dictionary.

I used the Collins Italian-English bilingual dictionary, which unfortunately has some serious problems. The upshot is that although it has a good selection of headwords, it does a miserable job of finding inflected forms. By contrast, the monolingual Zingarelli dictionary that comes with the Kindle for free has complete conjugations for all verbs. It also seems to have a good bit bigger vocabulary than the Collins. But, of course, being a monolingual, it gives Italian definitions for Italian words.

I ended up using the Zingarelli as my default dictionary. I would make a serious attempt to understand the Italian definition before giving up and switching to the bilingual. On the Kindle Voyage and Paperwhite, that's pretty easy to do. Here's an example from the first page. Suppose you don't know what soleggiato means. You press on it and get the following definition.

Between the context, "A magnificent summer day," and "well exposed to the sun," you probably wouldn't need the bilingual for this one at all. But assuming you did, the way you switch is you press on the words "Lo Zingarelli Vocabolario della Lingua Italiana" down at the bottom of the dialogue box. Kindle immediately pops up a dictionary selection screen:

Simply click on "Collins Unabridged Italian-English" and you're in business.

As expected, the word simply means "sunny." 

When Collins doesn't have the form, you work around the problem by opening the Zingarelli as a book and looking up the headword. For example, suppose you couldn't figure out riflettava

Again, the Italian definition is probably enough by itself, but if not, switching to the Collins gives you this disappointing screen:

So switch back to the Zingarelli and click on the "More" button in the lower-right corner of the dialogue box to get the "More Options" Menu.

Then tap "Open Dictionary." This opens the Zingarelli as a book.

Now we can use the Collins to help us read the Italian definition. In cases like this, the first thing to try is the headword itself. Press on riflettere and you'll see that Collins has the word after all. It just didn't have the form rifletteva.

Then just tap the "back" arrow at the upper-left, and the Kindle takes you right back to the novel.

I found that I was able to use the monolingual dictionary most of the time, especially toward the end of the book. A contributing factor is that authors tend to reuse the same words a lot, so if you have to look up a word at one point, you're likely to see that word again. If you made some effort to understand the monolingual definition the first time you saw it, then if you do end up looking the same word up again, the monolingual definition alone will usually jog your memory. That's a very good thing, because the more you spend immersed in Italian, the faster you'll learn.

I followed most of my own advice for learning a language by reading, and created flash cards for most of the words I had to look up. As a result, I learned about 100 new words in the course of the book.


Perhaps I'm getting better at Italian, but I found Sostiene Pereira an easier read than the two detective stories. Both of those contained at least a few sentences that I never managed to figure out, but I feel I understood every word in Sostiene Pereira. According to me, that's very satisfying.

Tuesday, October 28, 2014

Cinq semaines en ballon (Five Weeks in a Balloon)

On October 19, I finished reading Jules Verne's Cinq semaines en ballon, after ten weeks of on-again, off-again effort. Since the book itself came out in 1862, it seems a bit late to write a review of the novel itself, but it was the first novel I ever read in French, and since I've only been studying French for 40 weeks, I thought it might be of interest to discuss how I managed to read a French novel at all.

(All pictures taken from WikiSource.)

The Motivation: Reading on a Kindle

As I've mentioned elsewhere, I got excited about reading foreign-language novels about a year ago, when I discovered that I could use the bilingual dictionary on a Kindle to read a Spanish novel in four days. That was remarkable because I had spoken Spanish for forty years without ever managing to read four pages of a Spanish novel. I felt this proved the Kindle lets students beat the "beginner's paradox," and I wanted to get more people excited about it. Trouble was, my Spanish was too good in the first place--to be really exciting, this would need to work for people who didn't start off with C1 fluency in a language.

I had studied Italian on and off for about ten years, but I had never had better than A2 competence, so I brushed up my Italian for a few months and managed to read an Italian novel. As I expected, this was more work than reading the Spanish one, but still reasonable, and the effort improved my Italian dramatically.

But for a really strong test, I started studying French in January 2014, with the goal of showing that I could read a novel in a language I had never studied before at all. Moreover, I wanted to show that the mere act of reading could itself greatly boost one's ability with the language.

Elsewhere on the blog I've discussed my method for learning French. I use Duolingo as the core, and supplement that with other resources. I have invested about two hours per day for almost 300 days now. I estimate I'm somewhere between B1 and B2 French, which is about right for that level of effort. It's where someone should be after two years of college French, and those are exactly the people I'm hoping this will work for.

Reading the Novel

To read Cinq semaines en ballon, I carefully followed my own suggestions for learning a language by reading a novel. I was able to buy the original French text plus an English translation for under a dollar. The translation wasn't very good (it contained a lot of rather unnatural English sentences) but that didn't matter, since I only intended to use it as a last resort.

In a number of ways, this wasn't really the ideal first novel to pick. I chose it partly because I had really loved reading Verne novels as a kid, and this was one I'd never read before; I thought it would be cool to say I had read it in French. It was a little longer than I'd have liked--the first Spanish novel I read was about half as long. I also worried a little bit that French itself might have changed in 150 years, but that doesn't seem to have been a real problem.

A more serious problem is that like a lot of nineteenth-century literature, Cinq semaines en ballon starts slow and doesn't take off until you're more than 25% of the way into it. It's an adventure story about three men who ride a new kind of balloon across unexplored parts of Africa, but the balloon doesn't leave the ground for a long, long time. This is terrible for the first-time foreign reader, because those first chapters are unavoidably difficult, frustrating, and slow. It helps a lot if they're exciting, but that's definitely not the case with this book. From reading in the other languages, I knew that my reading speed would build with time, and that was motivation enough to get me through the first quarter of it.

After that, there's plenty of excitement.

A different kind of problem is that I started the novel before I finished learning the French grammar. I was comfortable with newspaper articles, but French has a verb tense called the passe simple which is almost unused in day-to-day speech but is very heavily used for narration in novels. It corresponds to the Spanish preterito and the Italian passato remoto, so I had a solid understanding of what it meant, but I still had to learn to recognize it. Since the narration is third-person, I only had to learn the third-person forms, and that turned out not to be too bad.

By the time the balloon finally lifted off, I had discovered another problem. The book is hopelessly racist. Given the era, that shouldn't be a huge surprise, but somehow I expected better of Verne. It's also not very environmentally conscious; the elephant in the picture above is eventually shot and killed, and the heroes eat just a single meal from a piece of the trunk. To be fair, there's not much else they could do given their circumstances, but neither the killing nor the waste bothers them one bit. This sort of thing doesn't turn up in every chapter, and it's not enough to spoil the book; it just jolts you every now and then.

That aside, it's a fun adventure with a satisfying ending. It turned out not to be such a bad choice after all.


My reading speed climbed from under 1% an hour to over 5%. It's hard to say what that means in pages or words, but toward the end, I could comfortable read two or three of the little chapters (44 in all) at a time.

At the beginning, when I found an unknown word, I consulted the bilingual dictionary first and used the monolingual only if that failed. By the half-way point, I had reversed this. The Kindle's built-in monolingual French dictionary turns out to be pretty good. Not only does it have a much bigger vocabulary than the Collins French-English bilingual dictionary, the definitions are written in simpler language. By the end of the book, nine times out of ten, when I needed to look up a word, the monolingual French dictionary was all I needed. It felt really good to stay immersed in the language and the story.

In the first half of the book, I made heavy use of the English translation. I always made myself read at least a paragraph before resorting to it, but all too often there was no other way to make sense of the text. Toward the middle, though, I was usually able to finish a whole chapter in French and then quickly reread it in English, looking for any big mistakes in my interpretation. There were quite a few to start with. For the last quarter of the book, though, the rereading didn't turn up any big surprises. That is, my reading in French was accurate enough that in the last quarter of the book, I didn't really need the English translation at all.

As with the other languages, I felt this really improved my ability across the board. For example, I'm comfortable reading most stories in Le Monde now without using a dictionary at all. I haven't attended a French meetup lately, but I should do that soon to see if there's a difference with my listening ability, as there was with Spanish.

All in all, it was a very worthwhile experience. I think I'll try Voyage au centre de la terre next.

Saturday, October 18, 2014

Flirting with French

I just finished William Alexander's wonderful Flirting with French: How a Language Charmed Me, Seduced Me, and Nearly Broke My Heart, and I highly recommend it to anyone with any interest at all in learning another language. In his thirteen-month quest to learn French, the 57-year-old author tries everything short of marrying a French woman, and he describes it all in hilarious detail. It's a story about a language, it's a story about people, and it's a story about not giving up.

I bought this book after reading an excellent New York Times review. Building on that, I'm going to go through the various things Alexander tried and comment on how effective I think they are for others trying to learn French. Since I taught myself French during the past nine months using many of the same materials and yet pretty much the opposite philosophy, I'm in a good position to compare and contrast.

Given the nature of his book, I don't think it's possible to have "spoilers," but it's possible some people might want to read the rest of this post after finishing the book.


What did he know to start with?

Alexander had studied French for two years in high school, and even though that was 40 years ago, it still seems to have given him a head start. He took an online college placement exam which said that he almost qualified for "entrance into first-year college French." Since the qualifications for entrance into first-year college French are tuition and a pulse, I assume he really means he almost qualified for placement out of the first semester. That's not bad at all.

Unfortunately, he hated his high-school French teacher and as a result had nothing but contempt for the whole concept of learning a language in a classroom. He intended to teach himself using the latest technologies, but without a teacher or even a tutor.

For my part, I had never studied French before at all, but because I speak Spanish and Italian, I had something of a head start of my own.

Why did he want to learn French?

Motivation matters, and his motivation was that he loved France and and all things French so much that he wanted to be French.  Obviously that meant learning to speak the language and not just read it; he wanted to attain conversational fluency in French.

My own goal for learning French was to be able to read books in it, and as a side-effect, to prove out my ideas that devices like an Amazon Kindle can make it much easier to read books in foreign languages.

What did he do?

In thirteen months, Alexander put 900 hours into active study, he spent "hundreds" of hours passively absorbing French movies, TV, and radio, he practiced with French people locally and online, and he spent at least five weeks in France. I estimate he spent at least $20,000 in the process.

If nothing else, one has to salute the power of his motivation.

For my part, I invested about 500 hours in active study. I also practiced with French people locally and online, and I spent about $100.

What was the outcome?

Using the levels from Common European Framework of Reference for Languages, I'd have expected that much investment to make him a strong B2 speaker, verging on C1 in some areas. Instead, he seems to have barely managed to reach B1. You can do a lot with B1 French, but given that given that he dreamed of being a C2 speaker, he missed the mark by a long shot.

I estimate that I'm also speaking B1 French, but I'm reading at between a B2 and a C1 level.

What went wrong?

Since we had very different goals, we took very different approaches. He shunned formal training, books of rules, tables of verbs, etc. Obviously those are things you cannot use in conversation; everything must come to you naturally. He was drawn to approaches with little or no written component, and definitely to those with no English component.

I bought grammar textbooks (my only actual expense), drilled vocabulary with flashcards, and used a free online program called Duolingo, which essentially trains you to translate from French into English. I used some of the other resources Alexander did, but only as a supplement to my main program. I do want to speak French (not just read it) but that's not my top priority.

Given, then, that we had very different goals, it's comparing apples to oranges, except for one thing: I think he picked the wrong goal in the first place. My opinion is that if you build up strength in reading and writing, you can leverage that to improve your ability to converse and to listen. Learning to read a language is not sufficient by itself to make you conversant--there are many, many people who can read English and even send e-mails but who can't utter a comprehensible sentence--but I think it's a necessary first step. This is backwards from how children learn, but adults are not children and do not learn the same way.

I should note that my view on this is rather old-fashioned, and a great many people don't agree with it. Alexander certainly didn't. Draw your own conclusions.

Tools for Language Learning

In this section, I'm going to go through all the resources Alexander writes about using, in the order that he mentions them. I'll summarize his results and then give my own opinions.

Rosetta Stone

What draws him to Rosetta Stone is the promise of teaching you the language "like children learn"--i.e. not with the methods of his hated high school teacher. Instead, it tries to immerse you entirely in French--spoken French--and get you to learn by osmosis.

Alexander's hilarious account of what it's like to actually use Rosetta Stone is in the chapter titled "First-Person Shooter." It's priceless. He never comes out and says so, but it appears that he got very little of value for all the hours he spent with it.

I am amazed that he had the determination to complete Rosetta Stone. Studies I've read suggest that almost no one does so. I won't go into depth on the problems with the package, as others have done so at great length. Since it's so far away from my own philosophy of language learning, I never even considered it, so I have nothing to contribute first-hand.

Vacation in France

A two-week bike vacation in France may not seem like a language-learning technique, but according to some research, a pending trip to a foreign country motivates people to do language study more than almost anything else, so as a technique, it's not to be underestimated. In terms of price, it certainly makes Rosetta Stone look cheap. That said, it's not clear how much you really learn on a vacation. You read signs and talk to waiters, mostly. A business trip where he'd have had to use the language all day long would have likely been much more effective.

I didn't visit France this year, although I have been there before, but all this French study has definitely given me a powerful desire to go back!


He spent an hour a day watching this famous international French cable network, but apparently never discovered that you can get it online. The free online version has an excellent set of lessons, which are targeted at making it easier for students to learn how to watch their shows, but he either didn't discover them or else decided not to use them.

Since he speaks about "enduring" the shows, I gather he didn't get a lot out of them. No surprise. Nation claims that you need to know about 6,000 word families before you can watch a movie, and that's two or three times what someone knows after two years of college study.

I made extensive use of the lessons on TV5Monde, even though they are really targeted at listening comprehension, because I didn't want to develop bad pronunciation habits. Their lessons are graded, and I advanced from A1 to B1 over the months. That means I can watch a French news broadcast and pick out a lot of it, but movies are still hopeless.

Online Immersion Class

Breaking a bit from his idea that he would learn French entirely through osmosis, he signed up for an online class complete with video and a live instructor in France. He appears to have done only a single lesson this because they expected such a high level from their students that English wasn't allowed even to ask for the definition of a word.

This is something I've never even considered doing. It's something that might make sense to try next year--once my reading is better. It does bring up another point; once you are strong enough in a language, you really can study it without using English. That is, people can teach you French in French. However, I think you need to reach level B1 before you do that. In my opinion, Alexander did this in the wrong order.

Fluenz French

He only mentions this $500 product briefly to share the fact that he didn't like it, although he later tells us he completed the whole thing. Fluenz is quite different from Rosetta Stone: It uses English in the lessons to explain some of the grammar, although it stops short of giving detailed rules for things like verb conjugations.

Alexander's complaints were more to the effect that he wanted more focus on speaking and that he found the videos distracting. It does seem that Fluenz should have been an improvement on Rosetta Stone, but apparently not much of one.

I hadn't heard of Fluenz before, so I searched the web for "Fluenz Problems." Apparently Fluenz only tries to bring people up to about the A2 level. Great for a traveler--A2 French is enough to get a lot done and it'll impress your friends and family--but it's not what Alexander was seeking.

Vocabulary sticky notes

I only mention this because it's something a lot of people talk about doing. Alexander and his daughter used sticky notes to label everything in the house in French. He only mentions it once, so I have no idea whether it worked for him. I have a suspicion that his wife didn't put up with it for very long.

I have never done this myself, but I can see the merit in it. Household vocabulary is important for reading, since many scenes in novels take place inside houses. You can also practice the language by making up sentences that use the words as you bump into them. The biggest limit is that it only covers a faction of the words you need--and none of the verbs or adjectives.


Frustrated with Rosetta Stone and Fluenz, Alexander decided to give social networking a try, and he enrolled in MyLanguageExchange. Oddly for a man so focused on speaking, he opted to use MyLanguageExchange strictly for e-mail correspondence with a pen-pal named Sylvie who lives in Orléans, France.

He used Google Translate, WordReference, and a printed dictionary to help compose messages, and he maintained a correspondence with Sylvie for months. When he finally met her in person, she was appalled at how bad his spoken French was, considering that his e-mails had been so well written. Alexander doesn't make a big deal out of it, but this is actually his one big success with French; he clearly developed decent reading and writing skills, even though that was not his goal.

There's a lot to be said for using social networking to help people learn languages. It lets you practice reading, writing, listening, and speaking with real native speakers. I practice writing using Lang-8, which is a bit more limited than MyLanguageExchange, but follows a similar principle. For example, I can write something in French, submit it, and get critiques from native French speakers. In return, I critique their attempts to write English. It's a great way to learn.


Alexander used Meetup.com to find a local group of French speakers who meet regularly for practice. He describes it as like attending a meeting of Alcoholics Anonymous, with twelve people sitting in a circle taking turns talking. Worse, his group had only a single native speaker. It's not clear whether he attended more than once.

It's a pity this didn't work out for him, because I've found Meetup.com an excellent way to practice conversation. Here in Seattle, there are French, Spanish, and Italian groups with a mix of learners and native speakers. There are weekly meetups, and when I go, I force myself to speak only my target language for two hours. If the goal is to be conversational, this is the best way to get there.

Memory Palace

A memory palace is the trick Sherlock Holmes used to quickly memorize lists of things. You visualize the words you're trying to memorize as lying in some physical location you're familiar with. Alexander says he successfully used it to memorize the entire 1,000-word children's English-French dictionary with 98.5% recall. Since the largest barrier to conversation for most intermediate students is lack of vocabulary, I really would have expected this to make more of a difference that it appears to have. Perhaps the memory palace is just too slow to use in conversation, but I would expect it helped him a lot with his reading and writing.

I've personally never managed to make the memory palace trick work, and I'm not sure it's really the best idea for anything other than lists that you need to recite in order.

French Movies

He watched at least one movie: Les herbes folles (Wild Grass). Since his wife watched it with him, it must have had English subtitles. I personally find it impossible to hear much of the foreign language in a film if I'm also reading the English subtitles. Subtitles in the foreign language work much better, but are harder to find. As I mentioned earlier, it takes a very big vocabulary to enjoy most foreign-language films. My bet is that he understood only a handful of words.


Reading is my favorite way to learn. Alexander casually mentions reading "the Respectful Prostitute" by Jean-Paul Sartre with the help of a dictionary and a translation. The "and a translation" is what worries me. It's very easy to half-guess at a paragraph, read the English, tell yourself "oh yeah, I got that," and keep going. You finish the book and imagine that you really read it. The same concerns apply to the dual-language book he read. Depending on how he read it, it might have helped his reading ability a lot, or it might have done little or nothing for him.

A disciplined person can make this work--someone who makes a detailed study of sentences they didn't get on the first try--but you have to really push yourself to do that. If you're enjoying the story and want to get on with it, it's hard to make yourself stop. If you're not enjoying the story, it's hard to do this at all.

I have my own ideas for learning a language by reading, and I feel I've been very successful with that approach. Given Alexander's success with his pen pal, I suspect he was more successful than he lets on.


On Skype, Alexander found a 50-year-old woman who wanted to practice her English in exchange for helping him with his French. Unfortunately, her English was so much better than his French that she couldn't even be bothered to try to correct his mistakes. This sounds like another one-time experiment, and, if so, I think it's another missed opportunity. Real-time conversations on Skype had the potential to make a really big difference to his conversation ability, especially since he had no luck with a local Meetup group.

Because meetup.com works so well for me, I haven't bothered with Skype, although I've thought of doing it to get exposure to more kinds of accents.

French Language Immersion Class: French on the Go

Eventually he broke down and decided to take an actual class. Run by the New School in New York City, this was a one-time, 16-hour immersion class with only four students (him included). The instructor wasn't a native, but that didn't seem to bother him.

In this class, he learned that he wasn't distinguishing nasal vowels at all. No wonder people couldn't understand him! I'm surprised he didn't find a way to get some sort of intensive training just in pronunciation.

French vowels have been, for me, the hardest part of learning French. French has lots of vowels that English doesn't have, but making the vowel sounds is only half the problem. English is rather sloppy with its vowels, and we tend to slur them all in the direction of "uh" in unstressed syllables.  To speak French, you must fight this habit and pronounce each syllable clearly and distinctly. It's like fighting with your own tongue.

Boot camp: Millefeuille, Provençe

Finally, he invested about $5,000 in a top-notch, two-week intensive-study program in France. This included room and board, 30 hours a week of classes (no more than four to a class), after-dinner talks, and group discussion. His description of his interactions with people during this period makes it seem as though he were actually speaking French at the B2 level or better. Given the level of individual attention that Millefeuille offers, I would have expected them to really work on his accent. He certainly finished up feeling really good about his French.

I did something like this for Russian back in 1992, when I spent six weeks in St. Petersburg. I do think two weeks is too little time, but I also think it was good that he did this last, since it would have been wasted had he tried it before he had a reasonable grasp of the language.

Real Life: Conversation with Sylvie

Remember the French pen-pal? After all this, fresh from boot camp, he met Sylvie and her fiance in Orléans for the very first time. It was a disaster. He couldn't understand their French, and they couldn't understand his, so they ended up speaking English.

This is why I evaluate his conversational French as B1. A B1 speaker can manage a conversation with people who are willing to indulge him, but the way he describes the afternoon, they weren't in the mood. A B2 speaker should have been able to manage it, even if it was a struggle.

Real Life: Bribing the Taxi

You'll have to read the book to learn why Alexander needed to bribe a taxi driver, but he managed to argue with the guy and get what he wanted all in French. For me this reinforces the B1 rating, since an A2 speaker would be unable to cope with a non-routine setting. At least, I hope it's not routine.

Miscellaneous Resources

At the end of the book, Alexander summarizes the resources he used during his 13 months of study, three of which hadn't been mentioned before:
  • Coffee Break French. These are podcasts, of which he listened to 100. It looks interesting, but I know nothing about it.
  • Pimsleur Audio Courses (2). I used Pinsleur Mandarin before a trip to China. It's an all-audio course, with no textbook. For my purposes it worked quite well--when I got separated from our group I was able to tell a Taxi driver how to take me back to our hotel. Pimsleur is generally thought of as a very elementary package, though, targeted at travellers.
  • French In Action (all 52 episodes). I like FIA. I haven't finished it, but it's fun, it's free, and it's 100% in French. It's an attempt to teach French in French, although there's a textbook that you can buy. As a supplement, I think it's a great way to get used to hearing French people speaking naturally but using few enough words that you have a hope of understanding them.

The Road Not Taken: Tools Alexander Didn't Use

Formal classes

"Classes are a sore point for me," he tells us, owing to that bad teacher he had in high school. He claims he couldn't find any nightly classes in his area and dismisses the idea of once-a-week classes out of hand--even though his wife claims to have learned Spanish that way.

I didn't use classes for French either, but I'm a linguist with an interest in second-language learning and I already spoke five other foreign languages. I think a good class (or at least a proper textbook) would have helped him a lot.

Alexander never mentions Duolingo at all. It might have had too much typing for his taste, or perhaps he thought it was too similar to Fluenz. It's also possible that he did most of his studies before Duolingo really took off.

I personally like Duolingo a lot, as I've described elsewhere. Duolingo is not a panacea, but I have to think that had he put hundreds of hours into Duolingo instead of Rosetta Stone, he'd have been better off. Richer too: Duolingo is free.


Perhaps he felt his memory palace eliminated the need for flashcards, but I think flashcards are the secret to learning lots of vocabulary, and I think Anki is the best flashcard program out there. The closest he comes to flashcards is his post-it notes experiment, and that's doing most of the work without gaining much of the benefit, in my view. Failing all else, I think flashcards would have been a good way to manage and maintain his memory palace, but he never mentions the concept at all.
Given that his focus was speaking, it's not a surprise that he doesn't mention this site where users help each other improve their writing ability. However, since it appears reading and writing are the two abilities he really did develop, it's the sort of site he ought to be looking at now. He seemed to have a strength there--perhaps not a surprise, since he's a writer. He should build on it.


Alexander's goals were different from mine. For me, reading is #1 and everything else is a distant second. For him, conversation was #1. Conversation practice is difficult to arrange, and the various programs that promise to "immerse" you really don't seem to deliver. Reading practice, especially using a Kindle, is easier than it has ever been. So I set myself a much easier task than he did.

Even so, he didn't really fail. B1 French is nothing to sneeze at, and, as I said elsewhere, I suspect he reached B2 or better in reading and writing. It's only that he set his sights so high and worked so hard that it seems like failure.

I do think he'd have gotten better results if he'd used his time differently. In particular, if the backbone of his study program had been a more traditional weekly class of some kind, then he could have augmented it with some of those other resources he used. If he had recognized that the written language came easier to him, and if he'd trusted that progress in that area would eventually help him with conversation, I think he wouldn't have wasted so much time bashing his head against the wall. Of course, that would have taken a lot of the fun out of the book!

I'm very, very impressed both by his dedication and his positive attitude. In that spirit, I'll close with Alexander's own words--my favorite passage from his book:
I may not have learned all the  French I wanted to, but what I did learn has enriched my life immeasurably. Yet perhaps the most important French lesson learned over the past year is this: you can love a thing without possessing it. Even as French has eluded me, my ardor for the language has only grown. I love, and will always love, French. Whether it loves me back, I have no control over.

Je ne regrette rien.

Sunday, October 12, 2014

Using Anki Flashcards for Vocabulary Drill

To read effectively, you need to learn thousands of words, mostly by rote. Flashcards are the tool par excellence for rote memorization, but, unfortunately, they become unwieldy beyond a few hundred cards. Anki is a free computer program that lets you comfortably manage the thousands of cards you'll need if you want to comfortably read books in foreign languages. This posting discusses how I use Anki, and shows you how to set up decks the same way I do.

What are Flashcards?

In its simplest form, a flashcard is a 3×5 index card with an English word or phrase (e.g. "to walk") written on one side and the corresponding foreign word (e.g. marcher) on the reverse. Make cards for the vocabulary you want to learn, shuffle the deck, and voila: instant random vocabulary quiz. Deal the cards one at a time and try to guess what's on the other side. Ones you get right go into a discard pile. Wrong ones you stick back into the middle of the deck. When you have discarded all the cards, reshuffle the deck and repeat until you can make a pass without any mistakes.

The astonishing thing about flashcards is how fast you learn. On the first pass, you typically get up to half the words because the simple act of creating the cards has taught you that much. On the second pass, that jumps to 90% or more. On the third or fourth pass, you get all or all but one or two, and by that point you're zipping through the cards just as fast as you can deal them.

Drilling with flashcards is fast, effective, and immensely satisfying. It's hard to imagine mastering a language without them. But you really don't want to do this with paper cards.

What's the Problem with Paper Flashcards?

If you just keep adding new words to the same deck, you run into a number of problems
  • Beyond one or two hundred cards the deck becomes impossible to shuffle.
  • No matter how fast as you are, a quiz with hundreds of cards takes a long time.
  • The deck fills up with words you know really well, and you start to feel you're wasting time reviewing them over and over.
You can avoid this by making multiple decks. That works to some degree for a language class, where you make a new deck for each new lesson and you don't review the old ones until exam time, when you dump them all together for a few marathon study sessions. If you do it that way, you'll find you retain some words pretty well, even after weeks, while you forget others almost at once. That may be okay for passing quizzes, but it's not what you need if your objective is to actually use the language. What you really want is a system that separates the "easy" cards from the hard ones.

In other words, you want the computer to organize it for you.

How does Anki work?

The Anki program lets you create decks on your computer and it handles running and scoring the drills. He's an example card from my English/French deck:

It's showing me the English side of the card and prompting me to type the French translation of "to walk" in the box. The phrase "move on feet" is a hint which I put there to remind me that this does not mean "walk the dog," which would require a different verb in French.

When I type marcher and press enter, Anki shows me the answer:

The green highlight on the word I typed means that my answer completely agreed with the expected answer. Below it is the content of the French side of the card plus the pronunciation of the word.

I can press any of the three buttons below that to indicate whether I got the word wrong ("Again") or got it right ("Good") or thought it was really easy ("Easy"). The numbers above the buttons indicate how long Anki will wait before showing me this card again. Cards that you get right over and over are rescheduled further and further into the future so you don't waste time going over them again and again.

Unlike a physical flashcard deck, Anki treats the French-to-English cards as being separate from the English-to-French ones. That makes sense because, for example, it is much easier to guess that extrême means "extreme" but much harder to remember that the French word needs a circumflex on the second 'e'. Whenever you do a drill, Anki mixes both kinds together. That is, some cards will show you English and expect French while others will show you French and expect English.

Anki is smart enough not to give you the same word in both directions on the same day, but here's what the French side of the card above looks like:

In this case the hint, "French," is there because otherwise I couldn't tell if this was the French verb "to walk" or the English noun for "a person who marches." I'll answer with just "walk" and not "to walk." Here's the result:

Anki shows me how my answer differs from the desired answer, but it still leaves it up to me to decide if my answer was good enough or not. In this case, I'd clearly mark it "Good" or even "Easy." Obviously that means I could just skip typing the words entirely, but I find that typing them keeps me honest. (I use the US International Keyboard layout to input characters that don't exist in English, by the way.)

Notice a couple of things about these cards: First, I only have to type the words, not the hints. Second, the pronunciation never appears with the question--only with the answer. That's very helpful because if it appeared with the English question, it would give the word away, but if it appeared with the French question, then I couldn't use it to help practice pronouncing the word. Always having it on the answer side means I can pronounce the French word out loud before I flip the card and then check myself. As always, it's up to me whether I count a pronunciation error as a serious enough error to call the card wrong.

Every day, Anki decides which cards you need to review (again, counting French-to-English as separate cards from English-to-French) and sets up a drill for you. It limits the drill to 100 cards (even if you skip a day), and you can stop in the middle of a drill and resume later (or not) if you need to. When you're done with the drill, every card will have a new review date, and Anki can show you some cool statistics to help you track how well you're doing.

For complete information, as always, read the manual. Anki has lots and lots of features.

How Do You Set it Up?

Download the app

The first thing you need to do is go to the Anki Website and download the program. Yes, this is a program, not a web application, so you actually have to install it on your PC. Don't worry--it's worth it.

Once it's installed, start the program. It should look something like this:

Yours won't say "Test Profile," most likely. There's no reason to name the profile until you want to have more than one, so don't worry about that for now. I didn't need a second profile until I wrote this article.

Creating a Deck

Downloading the Sample

The Anki User manual gives you lots of options for configuring a deck, but none of the defaults was quite what I wanted, and I went through a good bit of trial and error before I had something I was happy with. To make this easier for others, I have created a trivial sample deck. You can download it from here. The file called Sample.apkg is the one you want. If you've already installed Anki, simply download Sample.apkg and open it. That should cause Anki to import the Sample deck. Or you can just copy the file to your PC and import it manually. To do a manual import after you've downloaded Sample.apkg, run Anki, click on the File menu and select Import.

Either way, once you've imported the Sample deck, Anki should look something like this:

We've only downloaded the Sample deck so we can use the card type "Vocabulary," which is inside it. As soon as you have created a deck of your own, you can delete "Sample."

Creating your own deck

Click on the "Create Deck" button at the bottom of the dialogue box. Give your deck a reasonable name like "French Vocabulary."

You should see this on your screen now:

Click on your new, completely empty deck.

This is Anki's way of saying that you have no more cards scheduled to review today. Not a surprise, since you don't have any cards at all yet. Up at the top in the middle, find the "Add" menu and click on that. It should pop up this dialogue box:

You may need to resize it to get it all to show. Be sure it says "Vocabulary" in the upper left-hand corner (after the word "Type"). If not, click on whatever type is does say and select "Vocabulary" from the list. The whole point of downloading my Sample deck was to get you the Vocabulary type because it describes the layout of the cards and defines the fields.

So let's create a card. As you can see, we have five fields to fill in. Let's take them one by one.

Front: This is where you put the English word. I follow the rule that I always put "the" with nouns and "to" with infinitives, which helps me distinguish "to walk" from "the walk". Use a word from the language you're actually studying. This time I'll use "the leaf."

Back: This is where you put the word in the language you're studying. In this case, I'll put la feuille.

The next three fields are all optional, but when you need them, they're very useful.

Comment: This field that always shows with the answer but never with the question. I'll put the pronunciation here, but you can put anything you want there. I get the pronunciations from WordReference. In fact, to be safe, I like to copy the text from the Back field and paste it into WordReference just to be sure I didn't mistype it. That also tells me if the word has multiple meanings. Then I can click on the English word (still in WordReference) and see if that has multiple meanings. When there's ambiguity, I use the next two fields to clear it up.

FrontHint: This is shown with the English word, but it isn't part of the answer when you're drilling from French to English. I use this for three purposes: first, as in the "to walk" example, to clarify which meaning of the English word I want. "move on two feet" clarifies that I don't mean "e.g. walk your dog." For "leaf" I can say "e.g. on a tree."

The second reason I use a hint is when the target language has synonyms and I want to be clear which one I want. For example, cependant and pourtant in French both mean "yet" or "however". The difference is that cependant augments the first clause while poutrant contradicts it, but this is way too complicated to express in the hint. Instead, I just put "not pourtant" or "not cependant". The long explanation can go into the Comment field.

The third reason is when an English word and a French word have the same spelling and I need to specify that this is English. If the word and the translation are identical, I usually won't bother to create a card at all, so this doesn't happen very often. (E.g. "intelligent" in English is intelligent in French, so I don't bother to create a card unless I'm concerned about the pronunciation.)

BackHint: This field works just like the FrontHint, except that you see it with the French word. It's very important not to give too much away in this field. I want to memorize the French words--not my hints. I in this case, I feel safe saying "not sheet," and I add a bit more explanation to the Comment field.

Here's what it looks like all filled out. Again, don't bother with the three optional fields unless you need them; there's no reason to make extra work for yourself.

Now click the "Add" button. This actually adds two cards: an English-to-French one and a French-to-English one. In Anki terms we have created a single "note" which comprises two "cards." You create and edit notes but you drill on cards. If you realize you have made a mistake, you can click on the "History" button and it will let you edit the earlier notes (and this will fix both cards). If you look back at my examples of what a quiz looks like, you'll notice there is an "Edit" button in the lower left corner. That lets you edit notes in the same way even in the middle of a quiz.

Enter ten or so of your own vocabulary words and click the "Close" button when you're done. You should see a screen like this one:

Press the "Study Now" button and see how you do!

In a future post, I'll offer some tips and tricks to get the most out of Anki.