“It would take two lifetimes and more amphetamines than anyone could safely consume, to accomplish your first-rate ideas. That’s how energy-intensive they are.” — Khatzumoto
I came across AJATT about four years ago, in September 2011. In this site I found a guy eloquently and believably claiming that:

• You didn’t need classes to learn a language.
• You didn’t need to study grammar to understand and speak a language.
• You didn’t need special pedagogical (meaning: watered-down) content for learning the language.

What he proposed was rather radical:

• Find interesting content (both aural and visual) in your target language.
• Inmerse in it without understanding anything and let the process happen.

Before reading this, I knew from experience that what took me from “passable” English to fluency wasn’t classes (I had 14 years of classes), but reading interesting stuff in English. However, this guy proposed to practice inmersion solely and from scratch!
Khatzumoto eloquents talk about third rate ideas in the link I posted above. To him, learning a language with classes and grammar is a first rate idea: it sounds impressive, it makes a lot of sense rationally, but in practice takes up so much energy and effort that it is very hard to maintain.
In contrast, the passive inmersion method AJATT proposes sounds ridiculously easy: you don’t have to go anywhere, you don’t have to pay anything, you don’t have to study, you don’t have to take any exams, you don’t have to read or listen to anything that doesn’t interest you, you don’t even have to pay much attention. Despite the apparent laziness of this method, it empirically appears to work far better than classes.
This is why (good) third rate ideas win: because they take far less energy, and yet manage to deliver greater results.
Any self-improvement idea that appeals to my sense of laziness is something hard to resist. So I decided to follow his method and learn Russian this way. I figured it would be great to read Dostoyevski in the original, so why the hell not?
Fast forward four years. I have a roughly intermediate handling of Russian, both for conversation and reading – my grammar, particularly the written one, is dreadful. But I think I’m past the point of no return: in two years, or maybe in four, I will be able to read, write and talk in Russian with fluency.
What I want you to take away from this article are two things:

• The AJATT method works. This flies in the face of conventional thinking and 20th Century pedagogy. However, I suspect this method is close to how educated peoples learnt languages before the advent of mass schooling: by reading and dissecting the classics.
• The total inmersion aspect of AJATT turned out to be too much of a first rate idea for me, but I stumbled across an even lazier approach that seems to work for me.

In my AJATT process with Russian, I had four distinct phases:

1. September 2011 – June 2012: first love. I spent at least half an hour most days inmersed in Russian. In this period I learnt the alphabet, put my OS, Gmail and Facebook in Russian, started listening to Russian rock bands, watched Russian movies, etc.
2. June 2012 – July 2013: giving up. I went to Turkey and the desire to learn Turkish completely sidetracked my Russian. For a year it lay dormant and I didn’t do anything having to do with Russian.
3. July 2013 – January 2015: the bumpy phase. I started making an effort to read again every day, while trying to listen as much Russian music and movies as possible.
4. January 2015 – present: triumph of the third rate ideas. With minimal effort, and next to no guilt, I’ve managed to develop an intermediate command of the language. I also got the chance to go to Russia, which of course helped with the process.

The problems I had on phase 3, when I tried doing the AJATT method in earnest – but without the youthful energy I had when I started –  were the following:

• I wanted to listen to music that wasn’t in Russian, most of the time.
• I wanted to read stuff in English (both work and non-work related), most of the time.
• When I stopped using Google Translate (English <-> Russian), I would get tired after 10 minutes of reading.

In other words, the AJATT method was starting to feel a little bit like classes, in the sense of demanding too much energy and not being sustainable. In January 2015 I started experimenting with two third rate ideas:

• Listen to spoken Russian all day in the background at low volume, even while I’m listening to non-Russian music on top of it.
• Use Google Translate guiltlessly when I don’t understand a word in Russian.

To make the listening setup as lazy as possible, I downloaded an audio version of War and Peace, which is about 57 hours long (I’m not joking). Whenever I turn on the computer, I just turn on the book on a random spot, at a low volume, and leave it there for the whole day. I don’t pay attention to it at all and listen to music on top of it whenever I want. Yet, when I tune in to see what’s the narrator saying, I understand most of it. How did this happen? I don’t know.
Using the dictionary hasn’t hampered my reading, I think. After two or three lookups, 90% of the words will be memorized, because they have a rich context around them.
To sum up: passive, non-exclusive listening of the target language, plus using a dictionary guiltlessly, are taking care of my steady progress with Russian over the past nine months.
On top of that, I will sometimes do respectable actions by AJATT standards, like watching Russian movies and reading the lyrics of Russian songs. But for the most part, 5-10 minutes of reading a day, plus listening War and Peace on the background, are taking care of the job.
I encourage you to give this a try. I’ll be glad to hear any comments or questions you have about this. Write to me at fpereiro@gmail.com.