Why Language Learning Apps Suck – and Why You Should Use Them, Anyway

By OptiLingo

About ten years ago, I decided it was time to learn my ethnic language. In case you’re wondering, the language is called Circassian, and it’s considered “at risk” by the United Nations Educational Scientific and Cultural Organization. For what it’s worth, my non-profit foundation estimates that there are around 2.5M speakers of Circassian globally, with that number dropping by 50% over the next 20 years.

I’ll spare you the details of how (and why) I decided to learn Circassian, but a decade later, while I wouldn’t claim to be fluent, I can say anything I want to say, without pause or hesitation, and with a minimal accent.

I’ve also taught over 1,000 other people how to speak Circassian, in-person, over the web, and through the use of learning materials that I’ve developed over the years. Here’s a super high-quality picture of students being taught a course by me from a short while ago:

I share all this because I think it’s important to let you know who I am and where I’m coming from. I didn’t intentionally choose a click-baity headline, and I’m not pretending to be an expert when it comes to learning a new language.

That said, I am a guy who has dedicated a decade of his life to learning a semi-documented language with few resources, almost no television, radio or newspaper media, all while living in a country with only a handful of native speakers of the language in question.

I’m also a guy who’s put in this time and effort out of a sheer love for my heritage, and a guy who’s worked with over 1,000 people from the ages of 3 to 65+.


Like most people, I struggle to find time in the day

I have my full-time work schedule, I’m the father of three children, with a fourth on the way, and I’m always trying to find quality time with my wife, kids, family and friends.

Then I’ve got this other thing that consumes me… to improve my knowledge of Circassian and to teach it to my children.

Now let me just pause here for a moment and admit how crazy it is for me to spend my time on Circassian. There are around 5,000 – 7,000 languages spoken on the planet, and roughly half of them will die out in a generation or two. (My ethnic language, Circassian, may be one of them, by the way.)

According to David Crystal, in his book, “Language Death”, there are only a handful of reasons as to why someone would want to learn a new language:

  • For economic reasons
  • Because it’s the language of a state
  • For religious purposes
  • Because it’s the language of media

    The first three reasons make sense…. Many Americans learn Spanish for business reasons, and many Hispanic-Americans new to the United States learn English for the same reason. Icelandic is spoken by only a few hundred thousand people, but it’s the language of the state of Iceland, and Latin is still alive because of the Catholic Church. Finally, there are countless people around the world who speak English because they love watching Hollywood movies or singing along to American music.

    So why would I waste my time learning a dying language that’s not an official language and doesn’t live up to any of the standards above?

    I guess it’s the same reason that grown men (and women) still call their mothers and grandmothers on Sundays. There are many languages in the world, and mine isn’t any better than any other…. But it’s mine, and I love and cherish it.

    That said, I’m fully aware of how much time and effort it takes to learn a language- especially one that’s in decline, with few resources around it. So as a busy guy and a father looking to help his kids speak this language, I’ve been pretty interested in the best, most efficient way to learn a language for quite a while now.


    The Truth: Apps Suck

    So let me just come out and say it here to get it over with: language learning apps suck. They suck at teaching you a language. They suck at helping you to speak, read or write a language. But you should use them anyway.

    I know that’s a bold statement, so let me back that up with a personal story. In the Summer of 2016, I decided to visit a place called Abkhazia. Depending on whom you ask, this is either an independent nation-state with a strong relationship with the Russian Federation, or a renegade, break-away region of Georgia that is occupied by Russian forces.

    I’m not going to get into the politics of any of that, but I decided to go there because I have relatives who fled Syria to relocate to Abkhazia. (Side note: The Abkhaz and the Circassians are like Germans and Austrians; we’re a related people, but our languages are more distant than those spoken in Germany and Austria. Actually, a speaker of Abkhaz cannot understand Circassian and vice versa. That said, I have many Abkhaz cousins, and more than a few of them fled to Abkhazia when the war in Syria broke out. If you’re curious as to why or how an ethnic Circassian has relatives in Syria, then send me a private message, because it’s way too long of a story to get into here.)

    Now while most of my Abkhaz cousins actually speak Circassian, almost no one else in Abkhazia does. Also, as I noted above, speakers of Abkhaz cannot understand Circassian and vice versa. Because 45% of Abkhazia is comprised of non-Abkhazians, Russian is the dominant language.

    So I took it upon myself to learn Russian, and I gave myself about six weeks to do it.

    At this point you might be thinking that this is the place in the story where I tell you that I’m some amazing polyglot who was able to master Russian in a few short weeks using tools or methods “anyone can use!”.

    Or you might expect me to claim that I spoke amazing and wonderful Russian in just a few short weeks, or that I found the discipline to force myself to study every day for a few hours.

    None of that was the case.

    Frankly, I had these tools at my disposal:

    • Pimsleur
    • Duolingo
    • Memrise
    • Lots of my own materials

    For what it’s worth, I also have a copy of Rosetta Stone, but it’s probably the most worthless language learning product out there. I also have Michel Thomas, Living Language and Assimil, but I thought an app would be the better / faster way to learn.


    In a way, I was wrong

    I liked how Pimsleur was portable, and I liked how it forced me to speak, but it was a bit dull and slow. Also, as much as I love the methodology, I think that the actual dialogues and words are a bit dated. “Are you traveling with your wife and children, Mr. Pronin? Mrs. Pronin, I need to speak to your husband about work.”

    I definitely benefited from Pimsleur, and it was convenient, just not efficient, and difficult to stick with. So then I turned to Memrise. I figured I could augment my speaking ability from Pimsleur by adding a lot of vocabulary.

    After a hundred words or so, I had a hard time making sense of what I was learning. As with Pimsleur, I’m not knocking Memrise in any way. I think it’s a very well thought-out app. That said, what I realized was that learning vocabulary without context made it very difficult to really internalize what I was learning.

    Take, for example, the Russian words: пить and выпивать. They both mean “to drink”, but пить means to drink something that’s not defined, while выпивать means to drink something specific. That’s a difficult distinction to make without some explanation or a dialogue or sentence to put it into context. Now, granted, this is somewhat unique to Russian, since there are two versions of almost every Russian verb, but think about the English language. Imagine trying to explain the nuance of “however” vs. “but” or when and how to use “for” to someone without context.

    What I came away with was that flash cards (and by extension flashcard apps) are great for reinforcing bits of knowledge, but not very effective at picking up that knowledge in the first place.

    So I turned to Duolingo. Again, it is a very well-designed app, but it has one major problem: it doesn’t really teach you to speak a language.

    Let me qualify that. There are many, many things that Duolingo does well, but it doesn’t force the user to speak aloud on a regular basis the way Pimsleur does. It definitely put the things I was learning into context, and I think it helped me pick up more vocabulary, but it helped more with reading than anything else.

    Of course, when I landed in Abkhazia, I’d probably have to read a few things, but it was far more important to speak to the border guard than to read something he wrote.
    At this point, I’d burned through a week of my six-week deadline, and I wasn’t sure what to do.

    So that’s when it hit me… remember earlier when I said that I’d developed self-teaching materials for other Circassians to learn the language? (Roughly 80% of ethnic Circassians cannot speak our language. Again, long story….)

    Well, I had all of those materials cross-translated into Russian, Turkish and Arabic, since those are the languages that most ethnic Circassians speak natively.

    In the course of my decade-long efforts, I’d identified a list of approximately 900 core phrases, and an associated vocabulary list of around 1,500 words, broken up into three tiers, 500 each. (The first 500 words are the ones that make up those 900 phrases, and I have sample sentences illustrating the words in context for each of those 500 words, in addition to the 900 phrases.)

    I dusted off those materials and ran them through this website that adds stress marks to Russian words. This is very helpful, since stress on Russian syllables doesn’t follow any set rules.

    I then printed out my list of 900 phrases and the first 500 words (and their illustrative sentences) on an Excel-formatted sheet of paper. I also dumped the words (without the sentences) into Anki, and I used this site to download audio files.


    Beyond Apps: How I Taught Myself Russian

    Here’s what I did next, and the process was the same for the phrases, the words, and the sentences illustrating the words.

    • One by one, I would read 25 phrases in Russian, out loud, slowly.
    • I then looked at the English translation and internalized the meaning.
    • I’d then read each Russian phrase again, out loud, but more quickly.
    • When I had trouble with pronunciation, I would use back-chaining*.
    • Next, I would write out each phrase 2 – 3 times, depending on how much room I had.
    • Each time I wrote out a phrase, I would slowly speak aloud what I wrote.
    • When I finished writing out the phrase, I’d read it quickly aloud one last time.
    • I’d look back at the English translation and internalize the meaning.
    • I did this for all 25 phrases, then I went back to the top of the list.
    • With another slip of paper, I’d cover the Russian, look at the English, and try to recall the Russian.
    • I typically was able to recall 20+ out of 25 phrases without any difficulty.
    • I did this every day for my phrases, my words and my sentences.
    • This took about 45 minutes to an hour each day.
    • Each following day, I would review (but not write out) what I learned during the previous three days.

    I did all of the above in the morning, and each evening, I used Anki to review vocabulary.

    *Back-chaining is when you pronounce syllables from back to front. No one knows why this helps or works, but it does. Pimsleur makes extensive use of back-chaining, though the method predates Pimsleur.

    (Let me also take a moment to point out that this approach had some built-in self-reinforcement. The phrases I learned were comprised of words I studied on a stand-alone basis, and those words were given context through illustrative sentences that were different, but parallel to the phrases I was studying. It was almost like I had spaced repetition built into the DNA of the content I was learning.)

    So what were the results? Well, my Russian wasn’t pretty. My accent was lousy, and I made a ton of mistakes.

    But guess what? I was able to travel through all of Abkhazia on my own, without the aid of a translator and make myself understood.

    As ugly and miserable as my Russian was, the fact that I was able to navigate myself, understand what others said to me, and make myself understood for a full two weeks, after only about five weeks of study is a pretty positive outcome.

    That was last year. Since then, I’ve actually kept things up… off and on…. Using my own method, as described above.

    I remember when I was in college, I took German. That was about 20 years ago, and today, my German isn’t that bad, but back then, our approach was to go to class, study on your own and then attend drill sessions 2x weekly. We had a drill instructor who would use a form of spaced repetition to help us cement in what we’d learned in class and practiced on our own.

    I think that’s the best analogy for apps – they’re like digital drill sessions. They absolutely suck if you’re trying to acquire a new piece of knowledge (whether it’s language-related or not), but they can be very effective tools to help reinforce what you’ve learned through other methods.

    By the way, before I close this post, I should mention a few things: I am actually not a sample size of 1. Without realizing it, I’d used preliminary variations of my learning method with people whom I taught Circassian, and in retrospect, that’s probably why so many people I taught were able to pick it up. Also, since last year, I’ve tried this method out on a more formalized approach with a few dozen people, and the results are all pretty similar to my own.

    I should also point out that what I call “my method” is really just a combination of tools that are commonly used among proponents of “tactile learning”, which is a fancy way of describing “learning hands on” or “learning by doing”.

    Finally, I should also point out the combination of tactile learning approaches I use with my method are all supported by Hebbian theory. If you’re not familiar with Hebbian theory, you’re probably familiar with its most commonly used expression: neurons wire together if they fire together.

    My method forces me to read, write and speak aloud. The latter two are tactile learning approaches, but the combination of all three forces me to use different parts of my brain at the same time, which research shows is the fastest way to get things into long-term memory.