Category: Uncategorized

I produced a short film! It’s an interview with 6 native Esperanto speakers from different countries and they talk about how they wound up speaking this artificial language as their mother tongue and how this affects their lives. Fascinating!

(Be sure to turn on subtitles; the button is at the bottom right)

If you have more questions for them, they will be doing an AMA on Reddit later today, or you can ask on Quora.

When I was at the first ever Polyglot Conference in Budapest, there was an amazing lecture by Anthony Lauder. He revealed the mathematician’s approach to becoming a polyglot who speaks 9 languages:

In true mathematician’s fashion, he would then analyse the harder one of these two steps and try to simplify it. So: instead of “speak 8 languages”, we get “speak 7 languages, and add 1”. This can be further simplified: “speak 6 languages, then add 1”. And so on and so on, until you wind up with “speak 0 languages and add 1”. The audience was laughing. Anthony’s advice is as simple as it is effective: to become a mythical polyglot, all you need to do is Add 1 language – repeatedly, if necessary – until you reach your target. Don’t worry about learning 9 languages, just learn one more language than you know right now.

This home truth led Brian Kwong to create the Add1Challenge – a community effort to help everyone learn just one more language than they know right now. The challenge is open to everyone, no matter if they are learning their 5th language or their first, no matter if they’re starting from scratch or already know some. The key is to encourage each other, learn from each other and hold each other accountable, for 90 days. There is even a prize: the best participant will win a flight to the country where his target language is spoken. If that doesn’t motivate you, I don’t know what does. (Read more about the Add1Challenge)

Even as a veteran language learner, I personally found the Add1Challenge very useful. You see, I decided to learn Modern Hebrew and even made the mistake of promising Alex Rawlings that I’d speak Hebrew with him when he next visits Berlin – which will be on March 21/22, for the Polyglot Workshops. So I need to learn Hebrew quickly, and the Add1Challenge is helping me go further than I could without that community by my side.

This is me just yesterday, on day 21 of the challenge, speaking without notes (you can switch on English subtitles):

On Day 1, I only knew the alphabet and the two phrases “I am Judith” and “I am German”. I’m amazed at how much I’ve learned and looking forward to how much more I’ll learn before the end of the challenge!

If you want to try the Add1Challenge as well, you’re in luck, because the next batch is about to start. However, it is by application only, in order to ensure that everyone is serious about learning and speaking their target language. So go to http://www.add1challenge.com and click on “Apply” or “Notify Me” in order to get the chance to sign up for the next Add1Challenge.

Full disclosure: the above is an affiliate link, but I’m a fan of this challenge and would recommend it anyway. If you prefer a non-affiliate link, go straight to www.add1challenge.com.

Do you have questions about language-learning for which you’d like to know the answer? I just did a Hangout for participants of the Add1Challenge where they could ask all kinds of questions related to their learning.

Watch the recording of this video call by clicking on the video below.

Haven’t heard of the Add1Challenge before? It’s a 90-day challenge with the goal of having a 15-minute conversation in your target language at the end. I’m studying Hebrew and the challenge is helping me make incredible progress – will give you an update on Monday. Anyway, I recommend the Add1Challenge to anyone who seriously wants to improve. The next batch will start in a couple weeks. Check it out!

Full disclosure: if you do join the Add1Challenge based on a link from this page, I will get a bonus, but I would recommend it anyway, because I’m deriving so much value from it.

(Guest post by Chuck Smith)

Have you heard of the latest language-learning website that’s sweeping the globe: Duolingo? It was founded by Luis von Ahn, previously best-known for creating Captcha, those boxes on the Internet where you have to type a mangled word to prove that you’re human.

His next dream is to translate the entire Internet in every language, an ambitious goal to say the least! Through this new website, anyone can learn a language for free and once they progress far enough in learning the language, they can practice by translating real texts from their target language. Then people collaborate in a manner similar to Wikipedia and when the translations are good enough, they send the results to their clients and this is how they make money.

Duolingo and Esperanto

Luis von Ahn once said that Esperanto has been, by far, the most commonly requested language on their site as you can hear yourself from his lecture about Duolingo at Duke University. So, Esperanto was recently added as one of the new languages in its Incubator, which means that it’s open to collaborators to create the course. I was fortunate enough to be one of the first two people selected to start making the course from the very beginning. We were given a template for a generic skill tree with lessons to learn for English speakers, which we could modify as we liked to make it suitable for Esperanto.

One example of this customization is that relatively few Esperanto speakers know the word for suit (kompleto), because it’s fairly rare to wear a suit in typical Esperanto situations even though this is a really important word for other languages. In the end, we decided it wasn’t basic clothing vocabulary and moved it to a more advanced lesson called Business. Another example is that we’re working on a series of lessons called Affixes 1-3. That means that by the end of the course, you will have learned all the Esperanto affixes with examples of their usage.

In this system, we systematically teach vocabulary and every new word has to have at least three example sentences. With each sentence, we have to think of every possible translation. This makes sure you learn each word as used in a normal context and have enough reinforcement to help it stick in your head. Since we’re only human, once our course reaches a certain level of completeness, we’ll launch into a beta testing phase. Then the public can take the course and point out translations that are missing or mistakes in the course. After we’ve dealt with enough of these reports, then the course can enter its “stable” phase. At this point, Duolingo can start considering if they want to make other courses for Esperanto based on another popular language like French or Spanish.

Our team of five people has now brought Esperanto now over halfway through Incubation Phase 1 (course not yet released) and it’s been a wild ride. You can see all the courses in all phases of development here on the official page of the Duolingo Incubator.

The Future

I personally believe this will be known as one of the most important projects in the history of Esperanto. To get some idea of where Esperanto is going within the Duolingo system, it’s important to compare it with other languages within the same system. I would like to make comparisons between Esperanto, Irish and Ukrainian.

Ukrainian and Esperanto currently

First I would like to compare it with Ukrainian with the only reason being that the Esperanto and Ukrainian course description pages were both launched on Nov 4, 2014. Here people can sign up if they want to be notified when the course is launched. As of this time, 2610 people are waiting for Ukrainian (a language with 30 million speakers), while 6320 are waiting for Esperanto.

I personally believe that language course creators incorrectly estimate the market demand for Esperanto courses by looking solely at its number of speakers. In my opinion, this can be compared with looking at population density with respect to public transportation. Many people look at population density to determine whether high-speed train lines should be built. However, the correct question to ask when looking at market demand is the traffic population density. Many people have to travel between San Francisco and Los Angeles even though the people living between these cities doesn’t come close to the density of people living, for example, in many European countries. For more about this, see my article on Trains and Esperanto.

In the same way, it is more important to look at how many people are interested in learning Esperanto than its current number of speakers. In this sense, it’s interesting to note which language series offer a course for Ukrainian even though there is currently more demand for Esperanto: Rosetta Stone, Pimsleur, Colloquial, Pod101, etc. Some people may criticize these numbers and say that a lot of these 6320 are likely to be Esperanto speakers who just want to inflate the numbers, but I think a large amount of these are beginner or intermediate speakers who still want this Duolingo course to improve their language. A lot of people learning on Duolingo aren’t starting to learn their foreign language there from scratch.

Irish and the future of Esperanto

Another interesting metric is to take another small language like Irish. It is interesting to note that currently more people are learning Irish on Duolingo than there are native speakers of this language as posted by Luis von Ahn in this thread. I then answered this with my own parody thread that there are more people signed up for the Esperanto course than its number of native speakers.

So, considering that Irish has 130,000 native speakers with around 2 million speakers worldwide to various levels of capability, this is the closest language Duolingo offers for comparison to Esperanto to look into the future. The course went into beta on Aug 29, 2014, so now in a little under four months, there are over 250,000 people who have started learning Irish on Duolingo. This means that with a conservative estimate, it is very likely that next year over 100,000 people will have started learning Esperanto by the end of 2015.

Esperanto in Duolingo

So, last week we finalized our skill tree. You can see it below, but subject to change. Also, if you want to be notified when our course goes live, sign up on the official Duolingo Esperanto course page!

So, the most common question I get is when will the course launch? My best guess as of now would be May 2015. In any case, it will be very interesting to see what over 100,000 new Esperanto students does to its reputation around the world. Someone in a Duolingo thread did point out something else though, all these new Esperanto speakers will already, by definition, also be English speakers, which is true. But this could change when another Duolingo Esperanto course comes out, such as for French or Spanish speakers. Who knows what the future holds? At least as we go into 2015, we can be sure of one thing: “Roads? Where we’re going, we don’t need roads.”

Duolingo’s Esperanto skill tree so far

What would your requirements for a perfect language course be? I have spent a lot of thoughts on this question. If I could dream completely free from the constraints of existing books and audio courses, here’s what I would value:

• Highly interactive: rather than just listening or reading, I want to use the language myself
• Frequent feedback, so that mistakes won’t become fossilized
• Audio for everything, especially if the pronunciation of the language is very different from what I’m used to
• Useful vocabulary, not a single word that I couldn’t see myself using/needing, illustrated with example sentences
• Regular recurrence of vocabulary, so that it stays in memory – too many courses teach a lot of words and then never use them a second time
• Grammar when useful, for example too many students of German are forced to memorize declension tables when they still make many worse mistakes; the perfect language course would always consider what the biggest obstacle to comprehension is
• Plenty of exercises, because it’s always possible to skip some if I have mastered the topic, not possible to conjure new exercises if I’m still struggling to understand
• Clear goals and measurable progress towards them. Seeing progress motivates me.

When I was creating language lessons for GermanPod101, Myngle and other sites, I tried to incorporate these points as much as possible. I couldn’t change the format of the lessons though, so a lot of these issues remained.

Now I had the chance to design a language course from scratch, design a method from scratch. For the past year, I have been working on the language course of my dreams, the one that I wished I had had when I started out learning Chinese. Because it will Chinese. Chinese is generally taught so inefficiently, without regard to modern research on language acquisition, that I decided that improving the situation of Chinese learners would be the most worthwhile cause.

I am now ready to unveil my course to the world for the first time and I need your support to take it further.

Please have a look at LearnYu – the Automagic Chinese Course and then tweet about it and write about it on Facebook if you like the idea.

Here are some pictures to whet your appetite:

Please help me develop more lessons for LearnYu!

Thank you!

A lot of people refer to the Polyglot Community these days. I do, too, talking about the Polyglot Gathering and similar events. So who is this community?

It is certainly NOT just the 15-30 most popular polyglots who have Youtube channels or blogs. When I mean people like them, I personally use the term ‘Youtube polyglots’ or more broadly ‘internet polyglots’ – and by that I also mean for example Iversen of HTLAL or Simon Ager of Omniglot. My idea of an ‘internet polyglot’ is really anyone who is accepted as an authority on language learning by a reasonable number of people online, no matter what form of communication he/she uses.

The polyglot community however is much broader than just the ‘internet polyglots’, to the tune of tens of thousands if not hundreds of thousands of people.

Talking about it at/after the Polyglot Gathering, the consensus seems to be that anyone who loves to learn languages, even independent of how many language he has mastered, is a polyglot – it’s this attitude which sets our community apart from John Doe, who only learns a language because he has to. Love of learning and passion for languages.

Now there may be some hermits in Siberia who really love to learn languages without having any ties to any online polyglots. I probably wouldn’t include them in the term ‘polyglot community’, because there can be no community without being connected in some way to the other members of the community. But if you love to learn languages and have been to one of the polyglot conferences, you’re part of the community, if you’re a regular contributor to HTLAL or Unilang, you’re part of the community, if you follow at least one of the popular polyglot blogs or vlogs, you’re part of the community.

A small part of the polyglot community, at the Polyglot Gathering

Internet polyglots are not the whole of the polyglot community, they are the pillars. Polyglots come together around them and often because of them. Internet polyglots are also a representative sample of different philosophies, opinions and approaches to language-learning. For example, to start learning a language:

• Benny Lewis will start by speaking the language
• Steve Kaufmann will start by reading a lot
• Richard Simcott will follow a textbook or a class based on the communicative approach
• Professor Alexander Arguelles will look for a really old textbook based on the translation or grammar approach
• Niels Iversen will memorize ten thousand words from a dictionary

So there are many differing and even contradictory opinions in the community. There used to be quite nasty fights and vicious attacks, especially around Benny Lewis. These days, the disagreements are still there, but the attacks are gone (mostly), since people started to collaborate and to meet up in person. Even if they personally think that a certain approach won’t work well, people are happy to live and let live, let everyone pursue whatever approach he/she thinks is good, not rail against each other. After meeting each other in person or collaborating together online, most of the most visible members of the polyglot community have started to say:

“You know, I might not want to use your learning approach or methods, and I might recommend something different to people who ask me, but I recognize that we are all each of us helping people learn languages and I respect you as someone who is working hard towards the same goal I have.”

When you’ve finished a textbook, you’re often at a strange stage where other textbooks are too easy but real materials (books and TV shows intended for native speakers) are too difficult. At that stage, I’d sometimes use easy readers, but the stories rarely manage hold my interest. A better solution I discovered are bilingual books, which allow me to read interesting texts intended for native speakers while skipping past many of the difficulties.

Bilingual books have the same text in your target language and in your native language. There are several ways this can work, roughly from worst to best:

• Text in target language first, text in native language second. This is hard to use because you have to keep turning the pages in order to compare the two versions.
• Parallel texts (non-aligned). These have the text in your target language on the left side and the text in your native language on the right side. With this setup, it is already possible to compare texts more easily, but if the paragraphs don’t line up, it’s still not convenient to glance over if you need to know the translation of a word. You’ll probably just look at the English if you don’t understand several words in a single paragraph.
• Parallel texts (aligned). Same as above, but with paragraphs aligned, it often takes me less than a second in order to find the translation of the word I didn’t understand, so this is quite convenient to use. Various indie publishing houses offer parallel books and I have sometimes even created such parallel books myself (electronic versions only), either by hand or by using the handy software Hunalign. My friend Pete explains how to do it but you have to be a bit technical.
• Sentence in your target language followed by the same sentence in your native language. This concept, which was popularized by Franklang in Russia, also allows relatively quick comparisons once you’re used to it. I personally find it more disruptive than parallel texts though, because I cannot read only the target language, even when I’m in the flow.
• Interlinear translations. Here, the translation of each word is underneath the word itself, so that you can absorb new vocabulary and its meaning without having to glance anywhere else at all. This is clearly the best method if you want to learn a lot of new vocabulary while reading. It’s a lot of work to prepare texts in such a way, so they used to be quite short and rare, but now Interlinear Books has started to publish entire books treated this way. I recently had a chance to read the collection of Modern Greek short stories “The Clockmaster” by Roubina Gouyoumtzian, published by Interlinear Books, and I must say that they did a really good job. You can also read my review of this book.

So, which one do I recommend? All of them except the first. The thing is, while texts with interlinear translations are clearly the best, there are still very few of them out there. Parallel texts / books are easier to find, but still not exactly common. So don’t wait until you find the ideal method coupled with the ideal book that you always wanted to read; you’ll probably be waiting forever. Just go with what’s out there, create your own if you’re particularly tenacious, and otherwise focus on leaving behind that annoying intermediate stage as soon as possible. The real fun lies beyond, when you don’t need translations and you can simply pick up any book and read it with pleasure. And that’s what bilingual books help you achieve.

I studied Computational Linguistics, which is the part of linguistics that tries to teach computers human languages. It’s a very interesting subject, responsible for machine translation, dictation software, text-to-speech tools, dialog systems like Siri and many more applications. If you want to learn a bit more about Computational Linguistics, what problems there are for computers to deal with human language and what the future looks like, you can now watch my intro talk from the Polyglot Conference in Budapest:

If the idea of a Polyglot Conference appeals to you, check out the next Polyglot Conference in Novi Sad or the Polyglot Gathering in Berlin.

I’m often asked how many hours it takes to learn a foreign language, how many hours the asker should plan to spend on learning language X, (more rarely) how many hours it took me to learn a language. Here’s an attempt at an answer.

It is a difficult question because there are so many variables. You can get a first idea at Wikibooks:Language Learning Difficulty for English Speakers, because the language you’re learning (and the language(s) you speak already) are the most important factors. If you’re Chinese and trying to learn Dutch, you may need more than 2000 hours, while as a native English speaker you only need around 600 – and if you’re a native German speaker with some knowledge of English, you may be able to make do with 300 even!

So that table is not the end-all of estimates; it also depends on which languages you have learned already and if you have any experience at all in learning languages. The first foreign language is always the hardest, which is why Benny Lewis (author of “Fluent in 3 Months”) and I recommend spending at least 2 weeks learning Esperanto (which is completely regular and easy) in order to wrap your head around a lot of features that would be much harder to figure out when learning a less regular language.

Apart from your language knowledge and experience, the next biggest factor is the level you’re trying to reach. The numbers on the site I gave are all going for the lower end of professional working capacity. If you have a less lofty goal, you may be able to reach it faster, especially if you optimize for what you want to achieve. As you can read in this blog article I wrote for Benny’s site, I have been able to understand a Japanese TV series after 30 days, test at A2 level in Finnish after 30 days, give a public speech in Indonesian after 6 weeks, and so on. It’s all a matter of optimization for your goal. Few people really want to be an expert at every aspect of a language, at least not urgently. They urgently need one skill, but they develop all skills at once because that’s what textbooks and classes generally do. It slows them down a lot though. When I needed to urgently get comfortable reading Spanish in order to quote Spanish linguistics journals for my thesis, I reached that level in a few dozen hours of study rather than the 575+ that the FSI cites – but I don’t have professional working capacity in Spanish, I just learned enough for my purposes. It’s another example of the optimization I’ll do, accepting lopsided language knowledge in exchange for achieving surprising feats very quickly. Unfortunately, you don’t really get an idea of how to optimize your path until you’re already an experienced language learner. Here a coach may be helpful.

Note also that intensity is an issue. The FSI assumes that you will study very intensively, as their students do, at least 4 hours a day. However, the average adult language learner only studies two hours a week or so. At that speed, you will forget much more and waste time having to review it, so you will inevitably need longer than their estimates. The more intensively you study, the less total time you’ll need. I recommend intensive study especially at the beginning, to quickly leave the textbook stage behind and reach a level where “study” becomes self-motivating and fun, for example because you’re watching movies or reading interesting articles in your target language.

And that’s the real key: motivation. The more weeks and months you spend on a language, the more motivation you need in order not to quit. The vast majority of language learners quit before they reach their goal. So ensure that either a) the language is easy enough to pick up quickly or b) you’re studying very intensively or c) you have enough motivation to sustain you for the really long run.

Good luck!