Creator Resources: The Event Horizon

Olly Richards did a short video on the hyperpolyglot Vaughn Smith, who described how "every language makes people think differently. You realize there's a soul to the language, and you become another version of yourself." Another commenter (user "Cameron in China") noted that "In general, where Japan is minimalist, I find China can be quite maximalist." If nothing else, this observation does hold in regard to the language. Japanese has only about 100-107 basic sounds, which is very low compared to most other languages, and comparable to Polynesian languages like Hawaiian. Practically speaking, what this does is it increases homophony and the need for context-dependent communication. Recall that the point of learning a language isn't just communication. (If that was the case then simple AI translation would be enough.) The point, I maintain, is to inhabit the imaginative world of another cultural psychology, and at a still deeper level, explore the cross culturally evolving metaphysical narrative that is more fully revealed by means of immersion in another language and culture. (Per Tim Keeley, "language is a tool that directs our attention to various aspects of reality.") One may say that Japanese is unique in that it is both phonetically simple, and pictographically complex. But culturally, China and Japan do share a lot in common. I mean, if we go back far enough, "fossil and geological evidence suggests that Japan was connected to the Eurasian continent via the Korean land bridge around 80, 65, 43, and 20ka." Okay, yes that is a cultural eternity, but it's also the geological blink of an eye. Now, speaking of forming bridges... YouTube creator (handle: "kinda capable") gave his impression of using AI as a language tool. LLMs can be excellent conversational tutors. There's a lot of different takes on this, but I think he fairly represented a consensus opinion: 

"For a long time I was very skeptical of the answers ChatGPT gave me. AI tends to just fabricate things when it doesn't know the answer. And I was struggling to verify these answers. Although there's a lot of resources for Japanese online, it can be very challenging to find hyper specifics about the language. It's just so hard to search things in Japanese unless you already know Japanese. So, I started off using it sparingly out of fear of learning nonsense. But as time went on, I realized that its answers are, from what I can tell, very accurate on average. For some people, ChatGPT may be a reason not to try as hard. You know, it can just do a lot of the translating for you. But for me, ChatGPT seemed to be the best tool for learning the language myself, especially for free."

As the YouTube algorithm has guided me deeper down the rabbit hole of Japanese language learning videos, I watched a few more takes from less well known content creators. And some are rather good. Bunsuke, who just got a PhD in Japanese literature, took a low-tech, very systematic approach that relies on a lot of reading. Somewhat like Bunsuke, Chloe (who was born and raised in Australia, but has a Chinese background) took a traditional approach. She was also very focused on achieving popular benchmarks for success, describing how she passed JLPT N3 in seven months, and the N1 in two and a half years. I recall how a fellow classmate, Oleg Benesch, did a similar "speed run" ahead of his trip to Japan years ago. And recall that John passed the N1 in two years. Another creator (handle "BijuuMike") suggested using the app LingQ to support that method. He talked with Steve Kaufmann, who had this advice: 

"The fallacy of traditional language instruction is that you can deliberately learn something, that there's a direct relationship with your deliberate effort and what you end up being able to remember and use or recall. It's not that way. It all happens naturally if you expose myself enough to the language, either reading or listening and eventually using it. Deliberately trying to remember it leads to frustration. And I think that gets back to my sort of Zen, effortless, approach. You know, the best craftsman is the one who effortlessly (whatever he does, if he's a carpenter or if he's a butcher) just naturally does it.

There's a pattern in everything. There's even a pattern to English spelling. Everything has a pattern, and after a while you get used to these. And so it becomes easier and easier to remember new characters that you meet because there are familiar components. (One big advantage of learning Japanese and Chinese is that to me, although other people disagree, they are not grammar intensive languages like Russian, the Romance languages, or Arabic.) So you have to get used to thinking in terms of the patterns of the language rather than grammar rules. This is the way I approach Chinese and Japanese, "they say it this way." I wanted to get used to the patterns of the language, and I think that I carried that forward. So the first thing is, as difficult as it is, get the words in, get the patterns in, get some familiarity with the language, and then you can always go back later on."

Kaufmann leaves one with the very distinct impression that language learning is a combination of two methods, a combination of both effort and effortlessness, a coincidence of host and guest, of taking new information into our mind and also allowing our mind to be immersed within a new information environment, a coincidentia oppositorum. Returning to Bunsuke, he had some good book recommendations and a Substack. He shared a video with Joshua, the grandchild of a missionary to Japan (his grandfather was also interviewed separately). Since Joshua grew up in Japan, and Japanese is his native language, the effortlessness of his speech is apparent (Takashii and Unpacking Japan also interviewed him). Niko, the creator of "NihongoShark" describes language learning as a journey along which we try to reach a runaway chain reaction or self-accelerating state at which point progress becomes almost effortless. He used the metaphor of an "event horizon." If Steve Kaufmann had this metaphor, I think he would also use it to describe that point along a language learning journey we can all eventually reach. Here's Niko:

"Once you get that that good at Japanese, you're not going to forget it because life becomes intricately tied to Japanese. Many of your hobbies are in Japanese. Maybe a loved one speaks Japanese with you. Maybe it's part of your work. All these things happened to me so I'm not going to forget Japanese. To cross that line, to get past the event horizon you have to get a pretty high level of comprehension so that you can actually consume enough Japanese to make that enjoyable and doable. I'm not sure if people realize the way that this compounds over time."

Another content creator (handle: BJORN channel) said: 

“Stop trying to study Japanese efficiently. You can learn Japanese with whatever content you love. There's no need to have the perfect routine because if you're learning Japanese, you're going to be in it for the long game. We're talking years. And the method that's going to be the most efficient and effective is actually the one that keeps you showing up, learning every day. And that means getting lost in another world and forgetting about efficiency. Whatever world you want to be immersed in is the world that you can learn Japanese from.”

Maybe he was pushed to be more explicit, because he later recommended a number of free online tools, which he linked to in the descriptions of his videos. Definitely an involved process:

“Get this deck called "Core 2K" with sentences. It will teach the most common 2,000 Japanese words in the fastest time possible. The tool that's going to allow you to find comprehensible content is a website called jiten.moe ASB Player: This will let you turn any website into an interactive video player that saves audio, pictures, and sentences from anime, YouTube, or any other website instantly into a flash card. As a beginner, this is the primary software you'll be using. GameSentenceMiner: This will enable you to make cards from visual novels or games by scanning your screen. Ttsu: This will allow you to read Japanese novels and create cards. Jidoujisho: This is a full Japanese immersion suite that pretty much lets you mine from anything; it's for mobile. And last of all, the most essential application on the entire face of the earth for mining Japanese content is Yomitan: This lets you look up any word and get instant dictionary definitions to add to any of your cards. To put it simply, I just gave you every free immersion tool you can use to immerse in any media you could ever want. (And I made an entire separate video released simultaneously with this one that teaches you how to set up every software I mentioned in this video. I'll be giving you all of the presets I use, like making your Anki cards look nice, ASV player settings, literally everything.) I also convert everything I actively watch into audio files I can listen to throughout the day using this Audio Condenser Tool.”

Another creator (user handle: "danwashere") did the complete RTK method, then later reflected “In retrospect, I probably wouldn’t do RTK again. I'd have maybe tried to persevere with a vocabulary deck at the start.” However he noted how it was easy to use, and he could simply adapt the stories created by others. I thought back to the pop culture manga that Matt Alt mentioned, like “Space Battleship Yamato” or any other of the many different anime or manga, and imagined how that could make fun material to use as well. Anyway, his reflections: 

"Remembering the Kanji (RTK) by James Heisig was originally published in 1977, and the most up-to-date sixth edition was released in 2011 to include the additional kanji that was added to the Joyo Kangji list in 2010, because 2,000 kanji just wasn't enough. When I went on to YouTube and searched up how to learn kanji, there were many videos in which people pointed towards this book, but the one that caught my interest was by “Brit versus Japan.” He has a video in which he explained how he had learned over 3,000 kanji in 90 days through using RTK. He also has a guide on his website that he titled How to Learn Kangji Fast, the ultimate guide to remembering the Kanji. It is a very long guide, but I read through it all and just did what he recommends, which was to use a spaced repetition system such as Anki alongside RTK." This gentleman was also kind enough to have created a simple RTK flashcard deck. And so I imported that into Anki.

For the first few hundred kanji, Heisig writes out the mnemonic stories for you, but you don't have to use them. In fact, many people suggest making your own up from the start, but I found these to be helpful. He does this to teach you the technique of imaginative memory. But then he eventually stops giving full stories and only provides you with the keyword and primitives that are used. And from that point you have to make up your own stories, which if done with the necessary thought and effort does take more time. I would spend a lot of time trying to think of a story that includes the keywords and primitives. And sometimes this is difficult because of how random some of these words are. There are many people out there who are going through or have gone through this process. And there's a website called Kanji Koohii which was created to support learners who are doing RTK. There are plenty of community shared stories for each of the Kanji and and other users can even rate them based on how good they are. Now, I never actually used Kanji Koohii. Instead, I found a a handy little website that featured an RTK search engine which still included all of the community stories from Kanji Koohii along with a simple animation showing the stroke order. I can't remember how I came across this site, but after I did, you know, it soon became my RTK companion. I made an effort to not just rely on other people's stories, but used it as a a source of inspiration, seeing how other people had managed to form a story out of the key words. For over a month, I kept up with this routine.”

And what did I learn from all of this? I think it reinforces that Anki is a good approach, and there are a lot of way to make it more effective when combined with AI and comprehensible input. There are many entry points into this process, if I can manage to leverage them together to help make the journey both effective and enjoyable. Pragmatically, most appear to recommend taking a conceptually incremental approach to comprehensible input (text, video, etc.), however while advancing in each area simultaneously. And this also provides a method for reading textbooks. Learn how to use these resources, understand how to navigate around dictionaries and phrasebooks. I think Japanese language learning can begin with a global approach to every resource (or collection of resources), allowing you to mine or analyze it according to each of the following areas: 

1. Radicals (icon, index, sign) 
2. Characters (radical chunks) 
3. Vocabulary (character chunks) 
4. Sentences (lexical chunks) 
5. Scripts (sentence chunks) 
6. Books (script chunks) 

Building the Cathedral of Cosmoxenia: Getting into a flow state

The criteria for diagnosis with ADHD requires meeting a higher bar than most people who self-diagnose are actually able to meet, but that being said, I do have a restless brain that sits in a Goldilocks' zone on the "edge of chaos." All of which is to say that I function best when my mind is in a high speed flow state, that is, when it plays host to plenty of rewarding content to work with and time becomes less discrete and more fluid ("mental time travel," "sophisticated inference," "cognitive lightcone," etc). If I can provide it with high quality information to play around with and integrate into a broader gestalt, then everything tends to go well, problems become manageable, and insights flow freely. But the key is ensuring 1) that it has such information and 2) that this really is high quality content and not the empty calories of social media, pointless busy work, or simply stuff that was previously encountered. Now, I've had earlier successes with this method, which has uncovered some solid foundational work in philosophy and resulted in understanding other fields of inquiry, skills, etc, but those gains were not always easily reached. Taking this into new territory (that is widely perceived as difficult) will require reinvesting these earlier gains and employing some new methods to maintain progress. If the information I seek is either difficult for me to access or of low quality, I will not make progress. So I will do what is needed to maintain momentum with easily accessible, high quality, moderately challenging content. Toward that end, I will try to follow advice (use Anki flashcards and upload my own personalized spreadsheets) to propel myself into deeper engagement with the material. 

I also came across a language learning video (user handle "Jozen") which emphasized the importance of psychological performance hacks such as the Ovsiankina Effect, which was first described by Soviet psychologist Maria Ovsiankina in 1928. She demonstrated that individuals feel a strong, inherent urge to resume interrupted or unfinished tasks. This concept acts as a complement to the 1927 Zeigarnik effect, which describes how we are able to better remember incomplete tasks. The study was part of research at the University of Berlin in the 1920s that was influenced by Gestalt psychologist Kurt Lewin (see his work on friction and affordances within psychological "field theory," which decenters the significance of willpower). Starting a task creates a "quasi-need" or internal tension that persists until the goal is achieved. So one can easily see how this can be leveraged to increase productivity, though with the possible caveat that salience, temporal and spatial proximity, and other factors are all aligned appropriately. Notably, recall that "just get started" is Timothy Pychyl's most oft-repeated advice, and though I haven't seen him mention it, the Ovsiankina effect provides a technical explanation for one aspect of why just starting tends to work well. Correspondingly, it may also explain why perseveration on tasks we'd rather not be engaged with (and ADHD does lead to the rapid proliferation of these) can be very difficult to address. Essentially, just as the "helper effect" can indirectly lead to personal growth for the one providing assistance to a "guest in need," the Ovsiankina effect essentially encourages us to "help ourselves" to achieve the same sense of closure or resolution, as if we were our own guest. You don't need to knock down every domino, just the first one. And maybe a few more here and there, but it's the first one that's the hardest. (And it works just as well for supporting a healthy, active lifestyle, which is the foundation for success with language learning.) So let us begin. "A pencil! A pencil! My kingdom for a pencil!" As "danwashere" noted: 

"Mark Twain said "If it's your job to eat a frog, it's best to do it first thing in the morning. And if it's your job to eat two frogs, it's best to eat the biggest one first." The "frog" is the most important or difficult task that you have to do. And so, "eat that frog" means to do that task first before you get distracted or lose willpower. The first thing that I did each day was eat that frog, get it out of the way, and then I can go on with my day. I honestly do everything in the morning. That's why I like to get up so early at four in the morning."

Perhaps a union of the Ovsiankina Effect (just start, eat the frog in the morning) with the 80:20 rule (or Pareto Principle) can yield an effective strategy. You may have heard of hara hachi bu (腹八分目), a Japanese dietary phrase that translates to "eat until you are 80% full." Well, the 80:20 rule is a heuristic, a general distribution, not exact. There's many different descriptions of what it means and how it can be applied. One of these says that we should discard any heuristic we have that would suggest we can only act once we have arrived at a fairly comprehensive, or at least satisfying, relationship with whatever it is that is under consideration. Thus, any work ethic that appears to valorize the ideal of perfection should be carefully avoided. In place of that kind of approach we should instead be ready to move forward to the next step once we are only 80 percent of the way there. Whether that is 80 percent confident, complete, satisfied, or whatever. Here's the reason: getting to 80 percent only takes 20 percent of the effort, but finishing the last 20 percent takes 80 percent of the effort. Thus, if you only do 80 percent of everything, you are more than twice as efficient, and can get a lot more done. The obvious objection is that 80 percent of an airplane can't fly, so this doesn't make real world sense at all. The reply is that this only works when it is part of an iterative process. You do 80 percent of the work, then test. If you fail, then you do 80 percent of the remaining work. Test again. Eventually you will be able to move forward to the next step even though you assiduously avoided any illusions of perfectionism. 

Supposedly, advocates say that 20 percent of the effort normally put into tasks is really all we ever need, and that last 80 percent is primarily addressing our own emotional insecurities or tendency to needlessly perseverate. It's supposed to allow us to reach the same qualitative goals with less quantitative expenditures of time and effort, and thereby defeat Parkinson's Law, which is the boogeyman of all bureaucracies. Is there also a McGilchrist connection here? The left hemisphere loves perfection, of getting to 100 percent certainty, whereas to live life to its fullest we have to know when enough is enough. The Tao Te Ching says: 

"To hold and fill a cup to overflowing Is not as good as to stop in time. Sharpen a sword edge to its very sharpest, And the (edge) will not last long. When gold and jade fill your hall, You will not be able to keep them. To be proud with honor and wealth Is to cause one's own downfall. withdraw as soon as your work is done. Such is Heaven's Way."

It is best to live with some level of imperfection to avoid getting bogged down in diminishing returns and stagnation. One can learn to leverage that low level of discomfort to make consistent performance gains. Act as soon as one is just 80 percent prepared. We see this in sayings like "ship at 80, refine later," also described as "80:20 execution." The core philosophy is keep moving, avoid the diminishing returns of perfection, and engage in iterative improvement through real-world feedback rather than trying to anticipate everything beforehand. But we need to use these tools that leverage asymmetry in the context of higher-order volition (Harry Frankfurt), yet another asymmetry, if we are to pursue higher-order value (Max Scheler). Why? Because if all desires are thought of as existing on a level playing field, then we are extremely vulnerable to distraction. The attraction of relatively impulsive "first order" desires primarily lies in appearing to be simple to start, and simple to finish, while yielding an equivalent reward. They provide a "supernormal stimuli shortcut." And once engaged, stepping away from them means sacrificing that relatively uncomplicated sense of completion (Ovsiankina) and perfection (Pareto) they provide. But to reach higher goals, we may find that sometimes we need to make the hard choice to sacrifice acting on first-order desires for the sake of acting on second-order desires. Put the other around, most distractions implicitly require sacrificing second-order desire for the simple pleasures of a first-order desire, whether we realize that or not. In the moment it can feel very good to make that sacrifice. It's only later, sometimes days, weeks, or years later, that we realize the true cost. Scheler described the ontological depth; he explains what is higher. Frankfurt (famous for his book On Bullshit) gives the volitional mechanics; he explains how we can will the higher over the lower. And we may say that Charles Taylor, with his notion of “strong endorsement,” effectively fuses the two, in that the reflective structure Frankfurt analyzed presupposes an ability to endorse the vertical qualitative framework that Scheler described. 

These are all good - Scheler's higher values, Frankfurt's higher volitions, Pareto's power law distributions, and Ovsiankina's psychological tensions - but the motivation to eat the biggest frog doesn't come from any of those, at least not individually. That emerges from McGilchrist's process-relational, dual-aspect monism view of neurology: it comes from participating in a host-guest relationship. We get into the flow when we fully inhabit two separate roles, host and guest subpersonalities (see also Internal Family Systems Model). The host does its part. It can see those higher values and volitions on behalf of the guest. And the guest does its part. It can see those distributions and tensions on behalf of the host. Analogically, recall that the primary Bible verse where God decides to give man a helper is Genesis 2:18, which states: "The Lord God said 'It is not good for the man to be alone. I will make a helper [one who balances him, a counterpart who is] suitable for him'" (NIV). This passage suggests that Eve was created as a perfectly matched, complementary partner or "helpmate" to complete Adam. (The Hebrew word ezer is used here, which often describes a strong, essential partner, and is frequently used to describe God’s own role in helping people, e.g., Psalm 115:9-11.) Although, if we take a more figurative interpretation, then 'man' refers to 'human' more generally, and 'helper' refers to the left hemisphere, which per McGilchrist is the essential, complementary partner for the right hemisphere. Is this interpretation really supportable? Later in the same chapter in Genesis we read "and the two shall be one flesh." (Genesis 2:24) While the term 'guest' is ethically ambivalent, making it very useful for exploring the implicit tensions and potentiality here, a 'helper' refers to how that relationship with the host can manifest in an explicitly beneficial form.