- Lumi Lounge
- Posts
- Learn A New Language
Why learn a new language?
Why suffer through months or years of feeling like a child when AI is probably going to make it super easy to chat with anyone soon?
Ideal
I see two futures:
1. We wait. And, likely in a few years, we'll have some new headphones with AI that understands what anyone is saying and translates for you in real-time.
In this future, we wait. Let others toil to create a life of ease and comfort.
We see jobs whittle away. We hear stories of our friends and loved ones not being able to take of themselves because there's no work. We see new headlines of companies laying off workers. And we say:
"Ah, that's a shame".
Yet, we continue to wait.
Perhaps a better future is right around the corner.
Perhaps Congress will pass universal high-income.
Perhaps. We have to wait to see.
2. We act.
We see that the future is already rapidly forming. We know that there will be more technology, more layoffs, more technology, more layoffs.
We decide that we will take the wheel and guide the ship that flows on the river of fate.
We work to improve ourselves.
We accept hard new challenges and learn to surmount them.
We fight against the layoffs.
We don't just say we want universal high income; we demand Congress pass a robo-tax on corporations using AI and get it funded.
We learn languages to connect with other humans, to understand their culture, and to better ourselves.
In the future, we do these acts and more not, because they are easy but because they are hard.
The Obstacles
"Whoah, J, what are you talking about? I thought this blog was supposed to be able to learn more easily. Now, you're touting a hard narrative?"
1. The obstacle is me
Unfortunately, I recently realized that I may misled you. The point of Frictionless Learning is a preemptive solution to a real future pain. We're going to experience a massive economic labor market deflation.
My finger-in-the-wind estimate is that 80% of current active labor force workers will be displaced. At first, it'll be 80% of knowledge workers.
Then, with advanced robotics non-critical physical labor - like warehousing. Finally, critical trades-people like plumbers and electricians.
This is a matter of if, but when. Given the rate of progress -- even with accounting for limitations in chip manufacturing and the US electrical grid being able to support all this additional computing -- the “when” is going to be sooner than we think.
What to do? My solution is to get better at learning faster so we don’t feel like Baby Yoda.
We’ve all been Baby Yoda
We don't know the skills that the future will demand of us. We don't have the data to accurately predict what's coming in this domain. However, we can get better at meta-learning or learning about learning. And, we can use this same technology that is causing this pressure as a tool to assist in accelerating our learning.
AI-assisted learning can be:
- Personalized: AI will analyze your strengths, weaknesses, and learning style to create customized lesson plans just for you.
- Intelligent tutoring systems: AI-powered virtual tutors will be available at all times. They'll provide real-time feedback, answer your questions, and offer explanations -- making it feel like you have your own personal coach.
- Gamified and adaptive: AI learning systems will adapt to your level, and find the optimal point for motivation and difficulty as you progress.
- Contextualized: You'll be able to simulate real-life situations. Or, practice interviewing or foreign language conversations.
- Emotion-aware: AI will help you understand your emotional nuances and develop deeper self-awareness.
I don't just think this; I feel this.
Yet, I've been following a template laid out by other popular self-help speakers that advocate for doing ridiculous things like looking at your shoes for months when you want to build a habit of running.
If you want to start running, then start running.
Embracing that difficulty is what will defeat this first obstacle.
2. The obstacle is us
The size of the US gym industry is $31 billion USD.
The size of the US supplements industry is 193 billion USD.
That's almost a 4x difference.
The gym is a more effective solution for solving health and fitness problems.
Yet, we still pay more for supplements than we do for the gym. Supplements promise the same results faster and easier.
I'm not trying to change human behavior. I'm highlighting irrationality so that we can change our behavior.
Invest more in going to the gym than supplements.
Invest more in the hard path than the easy path.
3. The obstacle is our way of thinking
We tend to think from analogy.
Thinking in analogy is incredibly helpful for survival scenarios and in daily life.
For example: "Is that brown thing a bear?" | "Well, it's brown like a bear and big like a bear, so even if it is a rock I'm going to run away".
Or: "What should we buy at the grocery store?" | "Well, when I was growing up we ate baked chicken, collard greens, and cornbread. So, I'm going to buy the ingredients for that".
We compare our current problems, big and small, to other situations and can make rapid effortless decisions.
Thinking from analogy is incredibly problematic. Here are a few scenarios to exemplify:
"How should I learn Spanish?" | "Well my teacher said they learned grammar in school, so I should study Spanish grammar. I guess I'll go do that.” 💀
"How should we educate the masses of school children?" | "Well, this Henry Ford guy was able to produce a lot of cars quickly using this assembly line method. What if we applied that to teaching?” 💀💀💀
And, 100 years later the only Spanish we can speak from school are the lyrics from Despacito.
Uprooting the habit of thinking from analogy is very difficult to overcome. The only way I know to do so is to learn to think from first principles. I'm not going to explain thinking from first principles further because I've already written about it ↳ here.
It may seem like the work to overcome thinking from analogy isn't necessary. The reality is in most situations thinking from analogy is good. However, there are times when thinking from first principles is more useful. Learning a new language is one of those things.
The Path
Let's break learning a new language down.
A language has two channels:
Input
Output
And, has two forms (usually):
Spoken
Written
So, we can say that languages have four pillars:
- [Input | Spoken] Listening
- [Input | Written ] Reading
- [Output | Spoken] Speaking
- [Output | Written] Reading
I know you know this. But, breaking this down allows us to ask the critical question:
Which one of these pillars progresses language learning the most?
"J, what are you talking about - they're all important."
Yes, but one must be more important than the others.
Fortunately, we don't have to speculate we can use reason and research.
Stephen Krashen's research in language acquisition is at the forefront of linguistics. His research has two important findings:
Krashen's central theory is the Comprehensible Input Hypothesis, which states: language learners acquire a second language by understanding input that is slightly beyond their current level of proficiency (i+1). This "comprehensible input" allows learners to gradually build their language skills without excessive focus on grammar rules or error correction.
Acquisition vs. Learning
Krashen distinguishes between:
Language "acquisition" - a subconscious process similar to how children learn their first language and
Language "learning" - the conscious study of grammar rules and language structures.
He argues that acquisition, not learning, is the primary means of developing fluency in a second language.
"Wait, J, I thought we've been talking about language learning?
I used the term learning for convenience but let's focus on acquisition going forward.
If you want to acquire a new language, you should focus on maximizing listening and reading.
Yes. This is not what your Spanish teacher told you.
Stephen Krashen's theory has been battle-tested by poly-linguists, those who learn and speak multiple languages.
Online blogger Khazumoto - founder of AJATT (All Japanese All The Time) learned Japanese and got a job at a Japanese company in 18 months by applying Krashen's theory to Japanese.
Countless Youtubers also support this approach: Matt vs. Japan, Steve Koffman, XiaomaNYC, Ikenna. They’re all top-tier language learners and all advocate for this methodology.
"So, you're suggesting that all I need to do to learn a language is listen to it?"
Yes. I know it sounds too good to be true, but that is how we learned our mother tongue as a child. That is how I learned to speak Chinese. Even though, it sounds easy it requires listening to 100s of hours of content consistently. It's not easy, but it is easier than studying grammar rules.
Before you start binging Narcos allow me to give you some advice from the battlefield.
1. Reading is better than listening (or just watching TV). You should push yourself to watch content in your target language with subtitles on. Or, even better read a book or a short passage.
2. The content should be around your comprehension level. So, Narcos may be a bit out of reach for the moment.
3. Whatever you read or watch, it should interest you. When I was starting to read in Chinese I read a kid's book called "我的老师是火星人“ or "My Teacher is Martian". While I was reading content right at my level I found it dreadfully boring. This problem -- that of interest and level has been a thorn in the language learning community.
With Frictionless Learning I try not to just give you information, but also actual tools to be able to take on whatever you're learning.
I've spent untold hours learning to build AI tools and even more hours learning Chinese. A few months ago I decided to blend my interests together and build the best language acquisition app in the world. Sorry, Duolingo.
Lumi is an ai-assisted e-reader. You give it your favorite stories and it simplifies and translates that story to almost any language in the world. Then, once you open the e-reader. You can read while listening to the text.
It leverages all of the science we've talked about earlier and takes it one step further with unlimited personalization. Whatever you want to read, in any language, at any ready level is now at your fingertips.
I'm building this app to help you and me accelerate our learning using AI. To practice what I preach if you will. And, I've coded it 100% myself.
If you're reading this as a subscriber, you'll get 1 month of free access to the e-reader. And, if you're reading this and haven't subscribed, yet -- subscribe!
This is the first step. The app itself will change a lot in the upcoming months, so I welcome you to take the learning journey with me.
The way we live and work is rapidly changing. It's tempting to wait for technology to solve all our problems, but let’s embrace the challenge of learning hard things instead. Why a new language? Learning a language is not just about acquiring a new skill. It's about connecting with others, understanding their culture, and broadening your perspective – qualities that set us apart from machines.
Whether you choose to use LumiReader.ai or another method, I encourage you to take action today. Choose a language that interests you, find resources that align with your learning style, and commit to the journey. It won't be easy, but by investing in ourselves and our ability to learn, we are not only improving our own lives but also contributing to a future where humans and AI can work together to create a better world for all.
Learn. Step by step. Ferociously.
-Jeremiah
Joke of the week:
I heard Duolingo's owl is getting jumpy about these new AI language-learning apps.
It's been seen frantically repeating, "Hoot are you? Hoot are you?"
Subscribe if you enjoyed this article -- it's free. If you are already subscribed then share it with a friend!