There are two miracles that linguists study, although most of them don't call them miracles. The first is that I can stand before you and change the size and shape of my mouth and the location of my tongue as I exhale and I can cause a meaning to arise in your mind. And that meaning will most likely be very close to the meaning I intended it to be. I create a sound wave that meets your eardrum and then we are both thinking the same thing. How does a physical object, a sound wave, come to have meaning? This is the miracle of form and meaning.
The second miracle is that any human child can be placed in any linguistic environment, and she will learn perfectly the language of that speech community. At maturity, she will be able to understand and to produce an infinite number of novel sentences that she has never heard or thought before. She can do this because she has extrapolated the patterns and rules of the structure of words, phrases, and sentences in her language as part of growing up, and she knows how to apply those rules in new settings. And she can immediately tell if a new word or sentence doesn't belong in her language or dialect. Acquiring a language means acquiring a system that will generate all and only the pieces of that language. This is the miracle of language acquisition.
For example, you know immediately to make the nonsense word "breeve" plural by adding a "z" sound, but you will make "oik" plural by adding an "s" sound. And you know automatically that "Bnrtoud" is not a possible English word, just as "Himself kicked John" is not a possible English sentence, not one you would ever generate. Getting at what knowledge you have, what set of algorithms you have learned, that allows you to make those judgments is part of what the linguist tries to do.
We know that we do not learn sentences one at the time by imitation, since most of the sentences we produce and hear are sentence we have never produced or heard before. And we know that each language has an infinite number of possible sentences. You would need an infinitely large brain in which to store them. But even the patterns and rules we learn are not enough to account for one's adult competence. We know that you do not learn inductively because there would be an infinite number of lessons learned from inductive reasoning, leading children to form all kinds of rules that vary greatly from child to child, yet all children in a speech community zero in and speak more or less the same language or dialect. What guides them to the right target, the same target?
There is a paucity of negative evidence, that is, evidence of what is not in your infinity. It is easy to show that the data to which the child is exposed while learning her language is not sufficiently varied and robust to account for her mature abilities. The child must come biologically equipped to learn human language. This is called the poverty of the stimulus argument.
For example, the child learns to make the question "Is the man happy?" from the underlying proposition "The man is happy." The child also learns to make the question "Is the man going home?" from "The man is going home." We might think that the child has learned to take the first verb element, "is," and move it to the front to make a question. We already have a miracle. The child knows nouns from verbs and knows when to use one or the other. How did she learn this? No one gives a small child traditional grammar lessons.
But now take the question "Is the man who is wearing green happy?" This question comes from the underlying proposition "The man who is wearing green is happy." Note that "is" occurs two times in this sentence. Had the child followed the rule we formulated above, she would have moved the first "is" to the front and formed the question "Is the man who wearing green is happy?" but that's not the question at all. And kids don't try that and then have to learn better. They just know not to do it.
The correct question is "Is the man who is wearing green happy?" Kids who know English know automatically that in this case they will take the SECOND verb element, the second "is," and move it to the front of the sentence. They know that the question formation rule is dependent on the structure of the original proposition, and they know how to subconsciously analyze that structure and then perform the operations of their language or dialect. In this case, it means that they recognize that the first verb element is in a relative clause that it can't get out of. And being stuck in relative clauses seems to be a feature of all human languages, a part of Universal Grammar. Universal Grammar is what linguists call the inital state of the child's brain when language learning begins. Speakers know, therefore, to go on to the next verb. This shows that language is not linear but is structured in "chunks" in a hierarchical way and that you are wired to analyze language in a hierarchical way.
There is not enough data in the environment to account for the knowledge that the child acquires. Some of language is just part of being born human. Babies just a few hours old can hear the difference between "p" and "b". They don't have to learn this. They must learn, however, whether these sounds and the difference between them is used in their language to distinguish meaning, as in "pat" and "bat." The knowledge they have is a kind of innate knowledge that linguists call Universal Grammar that is encoded in your brain. It is part of your human genetic endowment that must be triggered and fed by your exposure to language in real life. It's what you start out with on Day One of your adventure in life and language. This is the miracle of language acquisition.
What is Universal Grammar? What accounts for the miracles? What parts of a language are native to that language and what parts come from Universal Grammar? These are the primary questions contemporary linguistics explores by comparing the ways in which the world's languages systematically resemble and differ from one another. I miss being a working linguist in retirement. Maybe some part time, maybe someday
