Luke Smith says that you will not learn a language by consuming media. Doesn't it contradict the AJATT theory?
There's this lazy idea that somehow if you passively sit around and watch people using a language this will somehow endow you with the ability to flexibly produce a language in the same way you see others using it. People want to believe it because they want to be able to watch TV or play a cell phone game like Duolingo or valueless Rosetta Stone-like software and somehow gain competence in a language.
It's not going to happen ever. Learning to play a boring computer game using words from a different language is not the same as learning to speak the language.
You might say of "just listening to speech" that "that's what children do," but that's not true at all. Children try pretty hard to participate and understand conversation. They sometimes have a desperate personal need to understand each passing sentence and hear the language they are trying to learn for hours a day for years. You watching some forgettable movie in the background as you play with your phone don't.
We talk about input and immersion a lot on this site. But the most important point to take away is that the input should be at least somewhat comprehensible and active attempts should be constantly made to make the unknown bits more comprehensible. If you just sit around watching something that you don't understand at all, don't expect to gain any competence in the language.
Some remarks about children: children learn to understand their native language first, and then learn to speak it. If you think about it, it's pretty obvious. If you learn a set of phrases in your target language, and then have a conversation with a native speaker, how are you supposed to understand what the person says back to you? As language learners, we need to practice speaking after we learn to understand.
In his post Luke talks specifically about Duolingo and Rosetta Stone. They are indeed very ineffective, but they have nothing to do with consuming media. Instead, they require active, deliberate yet extremely inefficient studying. We do not recommend apps like Duolingo as their methodology conflicts with AJATT's ideas of immersion learning. These resources base their instruction around constructing sentences from grammar rules, memorizing vocabulary in a specific order, and using questionable translations and low-quality text-to-speech software instead of audio recordings of native speakers.
Duolingo attempts to present itself as a game, yet those who have tried it know that it is not a very fun game to play. Duolingo has even implemented a notification system intended to prompt those who have stopped studying to come back. AJATT's concept of "consuming media" is different. It involves learning by reading, watching, and listening to authentic content in the target language, without relying on any phone apps. Anki is likely the only app one may need, but it is primarily a desktop app, with the phone version playing a minor role.
Luke also talks about watching TV or a movie in the target language. That's a completely different activity from using a phone app. First, I think, not so many people would actually watch a movie they can't understand yet without trying to make at least some attempts to understand what they're watching. You are watching a movie in your target language because it's fun, and you want to understand it. It makes no sense to watch it without trying to understand. You either watch a movie and try to comprehend it using a wide range of tools available to you, primarily dictionaries, or you already understand the movie pretty well, and you don't need intensive work to make it comprehensible. In AJATT we call the former intensive immersion, and the latter free-flow immersion.
It sounds like in his post Luke talks about something similar to free-flow immersion. During free-flow immersion you're paying attention to the content but don't pause, don't look up words and don't make flashcards. While it may not be as effective as intensive immersion in terms of learning vocabulary, it still has a positive effect on one's ability due to the fact that you can already understand most of the story, you are paying your full attention, and you can infer the unknown pieces from the context. Free-flow immersion is enjoyable and helpful, especially if you are an intermediate or advanced learner. However, it's not the only type of immersion you should be doing. You have to do intensive immersion to improve at your target language, but at the same time you need to acknowledge that it's not feasible to sit and do intensive immersion for 12 hours a day. You get tired of constant pausing and lookups. Free-flow immersion is a compromise between doing nothing and doing immersion.
AJATT also has passive (background) immersion. Passive immersion is when you listen to an audio recording in the foreign language while doing something else. Passive immersion is less intensive than free-flow immersion, is done only through listening and involves only partial attention. While doing passive immersion, usually you don't look anything up in a dictionary. Before immersing passively in a piece of media, AJATT recommends that one must first immerse in it actively. As an example, you could watch a Japanese movie and read its Japanese transcript (subtitles). Only then you extract the audio track to play it in the background. You need to learn the new words and grammar encountered during active immersion, as this will make it easier to comprehend the same piece when listening passively. When doing passive immersion, you're mostly letting what you've already learned sink in. You're not trying to learn new stuff. Again, listening to incomprehensible audio is not going to help you.
To summarize, what Luke says is mostly accurate, but he mixes together activities that are distinct from one another, and uses terms that have different meanings in the context of immersion learning. Sometimes he correctly identifies a problem but draws a wrong conclusion. Phone apps are not effective. Intensive immersion accompanied by using dictionaries and looking up words you don't know — works really well. Watching TV and movies in a free-flow mode is beneficial, but only when the learner is actively paying attention. Free-flow immersion becomes more and more effective as your level in the target language grows. Passively listening — works, but only if the learner has actively immersed themselves in the content before. Watching or listening to incomprehensible content — doesn't work or works poorly at best. It is, however, suggested that listening to incomprehensible audio has some positive influence on the ability to discern sounds.