Theory
Form and meaning
Second language acquisition (SLA) is a field of research that tries to understand how people acquire language proficiency. The recommendations made by SLA research are very different from what you will find in most textbooks and language classrooms. If you're skeptical of our advice to just 'read interesting books' or already have a lot of background with traditional approaches, it's important to understand the findings of this research in detail.
Here, we give a summary of the current theory on second language acquisition. Much of this page draws from Stephen Krashen's Second Language Acquisition: Theory, Applications, and Some Conjectures, which is readable enough that it can be read by non-specialists.
The acquisition-learning hypothesis
We have two very different ways of developing ability in another language: We can acquire language, and we can learn language.
Acquiring a language is unconscious. It happens when we're reading or listening to content that we care about. Since it is unconscious, it does not take active or conscious effort; it just happens. This is the way all of us come to know our first language.
Learning a language is conscious and willful. It happens when we study grammar rules, do drills and exercises, read a language textbook, or get our mistakes corrected by an expert.
These two ways of knowing a language are very different. Sometimes they are called 'knowing a language' (acquisition) versus 'knowing about a language' (learning). And sadly, it is possible to know a lot about a language without knowing it at all.
Contrary to popular opinion, acquisition is not just something that happens in young children:
The research strongly supports the view that both children and adults can subconsciously acquire language. Also, both oral and written language can be acquired.
The natural order hypothesis
We acquire (not learn) the parts of a language in a predictable order.
Acquisition occurs in a predictable fashion. Some aspects of language are acquired early; others are acquired late. And, this order of acquisition varies from person to person, although it's generally consistent.
Research shows that some rules that look simple are actually acquired quite late, and some rules that look quite complex are acquired early. Without detailed research, it is hard to know which is which. Further, research shows that we cannot circumvent this natural order through drills, exercises, and conscious teaching (learning).
It is not obvious which parts of a language are actually easy to acquire and which are harder. But even if these were known, each student has their own internal constraints on rule acquisition (an 'internal syllabus'), and these constraints vary from student to student. This suggests that a traditional 'grammar-first' language program is unlikely to succeed for most students.
The monitor hypothesis
Consciously learned language is only available to us as a Monitor, or editor.
Learned language, as opposed to acquired language, is available only under very specific conditions:
First, the specific rule must be known, which required deliberate study ('learning') as you will find in a traditional classroom or textbook.
Second, the user must be focused on correctness or linguistics form, as opposed to understanding and communication.
Third, the user must have enough time to consciously apply this rule. Normal conversation is too fast for most people, except for language experts who have already acquired the language to an advanced level.
Conscious learning is useful only for advanced language users who have the time and interest to monitor and edit. For beginning and intermediate users that have not acquired the language to an advanced degree, or who do not want to inspect the language in tremendous detail, such learning is likely to be of little value.
For Sanskrit specifically, the case for conscious learning seems even weaker. Most people learn Sanskrit so that they can read things written in Sanskrit, not so that they can produce correct Sanskrit in real time.
The comprehension hypothesis
Because of the importance of the comprehension hypothesis, we will quote from Krashen extensively (with emphasis added):
The Comprehension Hypothesis is the centerpiece of language acquisition theory. It attempts to answer the most important question in the fields of language acquisition and language education: How do we acquire language?
The answer is simple: We acquire language when we understand what people tell us or when we understand what we read. And there is no other way it can happen. While people differ in many important ways, they do not differ in the way they acquire language.
[…]
Here are two amazing facts about language acquisition: First, it is effortless; it involves no energy, no work. All that is necessary is to understand messages. Second, language acquisition is involuntary. Given comprehensible input, you must acquire – you have no choice.
[…]
If we provide students with enough comprehensible input, the structures they are ready to acquire will be present in the input. We don't have to make sure they are there, we don't have to deliberately focus on certain points of grammar. If this corollary is correct, it means the end of grammatically-based language teaching. It means the end of classes in which students focus on one rule at a time, "master it," and then go on the next. It means the end of boring lessons and texts, stories whose real objective is to provide practice with the relative clause. It means all that is necessary for language acquisition is input that is interesting and comprehensible.
Acquisition is due to input (listening or reading), not output (speaking and writing). Specifically, this input must be comprehensible, i.e. at a level that a student can understand. If these conditions are met, people will acquire a language and do so effortlessly and easily. Due to the size of a given language, we must create conditions for extensive reading.
The affective filter hypothesis
If the acquirer is anxious, has low self-esteem, does not consider himself or herself to be a potential member of the group that speaks the language … he or she may understand the input, but it will not reach the language acquisition device.
Any language resource should diminish affective variables by making students feel safe, included, and confident. In particular, many students find grammar-based language instruction to be boring, dull, hard, intimidating, or simply too difficult to understand.
The compelling input hypothesis
The compelling input hypothesis is almost like a counterpart to the affective filter hypothesis:
Compelling input appears to eliminate the need for motivation, a conscious desire to improve. When you get compelling input, you acquire language – whether you are interested in improving or not.
Any resource for language learning should be not just interesting but compelling. Since individual tastes vary so much, it is unlikely that any single book, resource, teacher, or author can be compelling to all students.