We can all learn grammar at any time we want. Grammars are readily available everywhere especially for widely studied languages.

It's a good idea to familiarize yourself with how grammar works in general without having to get a degree in linguistics. Having this foundation will help you a lot in learning foreign languages.

What I mean by "general" is not the actual details of learning a foreign language. Because when you focus on the details, it's like seeing the leaves instead of the forest, and you may be disillusioned by why you're actually focusing on certain things and how they fit into the bigger picture of things.


For example, if you're working on the conjugations of French imparfait or Russian perfective verbs but you don't have a slot in your mind of when you actually need to use these tenses, then you're just adding useless trivia that you have no use for. You're memorizing things for the sake of memorizing rather than actually improving the way you communicate.

If you're doing this for cross-language comparison or research, then you have a specific goal for looking at this data and it makes sense to you.

Otherwise, you need to take a step back and take a look at why this grammar exists. And that's where we should begin our discussion. We'll define what your timeline for efficiency is.

First, what are your goals for learning a foreign language? Is it a pastime that you enjoy and figure this is something you're going to do casually for decades to come? Or do you have a time schedule, to reach fluency by a certain date, start using the language functionally for work or school or for some personal endeavour?

Second, how deep and what aspects of the language do you want to learn? Is it for reading literature in the native language, for translation, or for professional capacity? In other words is your goal deep mastery (C2) level, or is it for casual conversation (A2~B1) or for business or full native functional conversation (B1~B2).

If your purpose for learning is casual without any specific goal in mind, then reading about grammar and doing grammar exercises for the mere pleasure of it is fine. The more you're exposed to it, the more you'll know.

However, if you're on a tight schedule, it's important to figure out when to focus on grammar for best results in achieving your fluency goals.

The Child's Development

When a child is born it cannot see, feel, hear, or touch. Not in the sense that we do. Every single environmental input has to be learned and decoded. We know that when blind people who have been blind from birth get sight for the first time, it also takes them many months of training to learn all the fuzzy, blurry colors and shapes in the world and finally learn how the eyes focus to see clearly. I don't know for sure, but opening your eyes for the first time must feel intimidating, scary, and maybe even give you a headache because of the overwhelming amount of new data coming in. The same can be said of those born deaf who hear for the first time. They would not be able to understand language, and would have to start from scratch trying to decipher the sounds, and it would be very difficult to actually acquire the ability to speak if such an operation occurred after childhood. It would have to be partial deafness or blindness that would go a long way towards acquiring normal functionality.

Over time blurriness turns into clearness. And from clarity, finer distinctions can be made. So you've learned all your colors, but now you need to make a finer distinction between two shades of red. Everybody in the world can do this regardless of their language and they find ways to communicate the shades when they need to. Just because Russian has a word for dark blue and a word for light blue and English doesn't, doesn't mean that English speakers can't see or even express that difference (I just did), though there seem to be quite a few misguided studies and overzealous journalists of sensationalism in need of spreading clickbait.

If we observe children and their stages to fluency we can learn a few things. According to studies at Stanford University, between 12 months and 36 months of age the child acquires a functioning vocabulary of about 600 words, and during that process some of the last words to be acquired are pronouns.

Some of the first words a baby learns are names of things; really important things in the child's life like foods or toys. Slowly, abstract things start creeping in like describing things with adjectives -- though some children are slower at grasping the abstract aspect -- all children eventually do and this abstractness makes human communication so amazing! So a "ball" can now also be called "big" or "small" and these are labels that lots of different things can share that don't actually change their name (noun, that is). In other words, a "ball" is not a "dog", but both can be called "small", and that is something powerful that makes human language full of infinite possibilities.

Starting from personal needs, the child starts to learn verbs like want, not want, hungry, thirsty, fall asleep, but then also fun action things like hit and kick and run.

And slowly but surely all these categories of word groups grow and grow.

At first, anything that moves on wheels or that looks big and moves is a "car". Slowly but surely, finer distinctions start to be made as time goes on and the keen child will even start recognizing different car brands. Most children feel no need to take it that far until they're teenagers. So this is one way of measuring curiosity and intelligence in children as their thirst for knowledge exceeds their own hierarchy of needs.

The same is true in all areas of science. Something that you call a "flower" is absolutely not a flower to a botanist, but any one of hundreds of possibilities. The more you're interested in a specific topic, the more specialized your vocabulary becomes.

So after a while, the child realizes that a car has wheels and doors and other things that look different in the car shape away from the car. The child realizes the house door and the car door are quite different! And the brain makes great new connections telling the child that similar shapes that have similar functions can be called the same thing, but not all things that look alike can be called the same thing. And this is an amazing thing that humans can apply to constructing the myriad of meanings that our languages carry but most of the more advanced area of vocabulary expansion happens subconsciously. Once the child finds a truth in the world, the brain starts applying that truth subconsciously to every new pattern it comes across.

Since we understand the sequence at which children are learning words and putting sentences together, we can also deduce how grammar creeps into their sentences and starts to sound like cohesive units.

Before You Start Learning That Language

Get familiar with the language you're learning and see if you can answer the following questions:

What language family does it belong to?

We should know this because it tells us how much work will be required to learn vocabulary. So since English belongs to Indo-European, then I know that any other language within Indo-European, from Bengali to Armenian to Icelandic, is going to share roots with English, no matter how different they now appear, the process will be easier and word construction will be easy to decipher. If a language is not Indo-European and I don't previously know any language in that family, the process will now have become much harder and a lot of effort will be required to acquire vocabulary.

What we now call Proto-Indo-European is a language that was spoken long ago. It is like the great-great-great-grandfather of the whole family, and all of today's modern languages are his great-great-great-grandchildren. They all share the same DNA even if they look and sound different today. We're all a family. Some families are very big, others are very small.

Knowing some simple sound correspondences can help me pick up lots of vocabulary. Knowing that the ending of "weight" has fricatized into "-s" in another language helps me remember Russian "ves" or Spanish "pes-" much easier, not to mention the beginning sound is a bilabial in all three with more or less the same vowel in between. Hindi "vaj-" and Persian "vaz-" are close matches despite the languages seeming to be farther removed.

Can I say all the sounds of the language?

Sounds often have little to do with writing systems. Writing systems are just convenient ways to encode sounds ad hoc, and sometimes meanings. I like to call writing systems the "lossy format" of spoken language! For example, English "way" and "weigh" have different shapes but sound the same, so they encode meaning just as much as they do sound, which is loosely related in the spelling.

Learning the IPA points of articulation can help most of the problems and being able to control the tongue to produce each point of articulation. Can you unaspirate your stops (Italian), aspirate them (English), pre-aspirate them (Icelandic), release them tensely (Korean). Can you make voiced and unvoiced stops (p, t, c, k) at the bilabial (everybody), interdental (Icelandic), dental (Spanish), alveolar (everybody!), post-alveolar (English), retroflex (Hindi), alveopalatal (Russian, Chinese), palatal (Hungarian), velar (everybody!). Can you make each of these into fricatives (continuous sound like [f v s z ʃ ʒ ɕ ʑ x ɣ]) and into affricates (stop + continuous like [pf bv ts dz tʃ dʒ tɕ dʑ kx ɡɣ]).

If you can do just that, you've come a long way and most languages won't give you too much trouble. There are trickier things going on deeper in the mouth and throat, and some other special combinations with laterals ('L's), but the previous paragraph should help you pronounce over 90% of the sounds in the world's languages.

Vowels can be tricky, but even if you can't get them exactly right, remember that among hundreds of thousands or millions of native speakers, vowels happen within a range so that if you get somewhere inside that range and your other vowels are distinctly outside that range, you'll be understood. Your own vowels shouldn't overlap like a Venn Diagram even if they overlap with another speaker's vowels because their ears will adjust up or down to your positioning.

We have published a lot about pronunciation (download some of our ebooks covering pronunciation in various languages), the IPA (check out our Glossika Phonics channel on YouTube) and we have more coming out all the time, so don't forget to subscribe to our newsletter here.

What is the Structure of the Language?

Normally we talk about languages being Subject - Verb - Object (SVO) or any combination of those orders: SVO, SOV, VSO tend to be the most common.

This idea works great for most of the pan-Eurasian languages. But not all languages on earth follow a subject - object paradigm, which is something I'm starting to write a lot more about. Sometimes the relationship between the subject and the verb is vague which is why when you say "an accident happened" the accident didn't do anything, and it wasn't acted upon by something else, it just came into being during the course of the verb event. In English what appears to be the "subject" comes before the verb, but in both Russian and Chinese it comes after the verb, though in Russian "avaria" (accident) is still unmarked as "subject" and Chinese never marks parts of speech, though as a speaker of Chinese, I'd say it feels like a "subject" in Chinese.

The reason for this is due to the fact of how we can classify verbs. And if you're learning far-flung languages, it's really important to learn how to classify verbs at the topmost-meta level as you can.

Many verbs that involve the body and its functions (sleep, burp, cry, sneeze, etc) are not really voluntary and in many languages these verbs simply do not have Actor-Subjects (in other words, I/you/he-type pronouns). Instead, they are perceived as happening to oneself uncontrolled, and therefore use me/you/him-type pronouns!

But children learn this from Day 1 without thinking about it. It is ingrained from the very beginning. How? Think about it... Pronouns are the last thing they pick up during their first 36 months, and by that time, the actions that are perceived as happening to me are automatically attached these types of pronouns.

The foreign language learner, by contrast, is stuck in grammar hell buried under tomes of grammatical tables and convoluted explanations.

Please, I implore you, use a top-down approach to grammar, then start adding the details later -- when you actually need them. Don't be the zealous child trying to remember car brands. Be the teenager who needs to know them because it's become part of your world. (Hey, sometimes it pays not to be a genius!)

Four Verb Categories
  1. Existential: Have-Ownership (person to thing, person to person kinship, thing to thing part like my car has wheels), Existential (there is/are/was/were, happen, occur, etc)
  2. Stative: Location (where at, where to, where from), attributive (adjectives like be hot/cold/hungry -- yes you conjugate these with the be verb so they're actually verbs), and everything emotional covering cognition, desire, emotion, habit, personal experience, opinion predicates, perception, and proposition.
  3. Active: these are all the action verbs that have true subjects. Don't confuse these with Causative verbs though. Many action verbs that take objects are more likely to be causatives because it undergoes a change of state like explore, tear, break.
  4. Causative: these verbs are hard to identify for most people who don't speak languages that make them obvious. Languages like Japanese make them pretty obvious and it's a point of difficulty for learners.

Common causative verbs published in our Word Order language books: accumulate, add, affect, afraid, assemble, betray, boil, borrow, break, brush, build, burn, change, close, combine, complete, complicate, connect, conserve, construct, convert, cook, correct, cover, create, cure, damage, deafen, deposit, destroy, develop, dilute, diminish, make disappear, disassemble, disconnect, disperse, disprove, dissolve, dissuade, disturb, divide, get divorced, earn, enlarge, erase, establish, exchange, exclude, expel, fasten, feed, fill out, fix, fold, force, fry, harm, hide sth, hinder, hire, import, include, increase, inherit, install, interest, interrupt, intimidate, invent, involve, irritate, kick, kill, lend, liberate, lock, lose, lower, maintain, make up, manufacture, melt sth, memorize, mend, mix, move sth, offend, omit, open, organize, paint, plant sth, postpone, pour, preserve, press, print, produce, provoke, purify, raise, recover, reduce, release, remind, remove, repair, resume, ridicule, rouse, ruin, save, scare, sell, separate, set up, shave, shorten, show, make sth slow down, smash, smear, sort, spend, spill, spoil, spread, stimulate, make sth stop, straighten, make stronger, teach, tear up/down, throw, tie, make tired, torture, trade, transform sth, translate, unlock, untie, violate, wake sb, wash, waste, wipe, cause worry, wrap.

Once you master these categories, you'll need to ask the next few questions:

Does the language have different nouns like we do for pronouns I/me/mine etc.?

The next question you'll want to know is:

Does the language mark nouns differently based on whether they're active or causative (like Japanese -wa and -ga)?

Knowing the four verb categories above will help you group things together properly.

Does the language use prepositions or only postpositions?

The secret is that postposition languages are actually easier: Finnish, Estonian, Hungarian, Turkish, Kazakh, Mongolian, Korean, Japanese.

The preposition languages that also have nouns change for declension means that you have to remember two things for every role in the sentence. Not only that, but prepositions are the most irregular nonsense words humans have ever invented. Learners of English are lucky that they don't need to decline English words, but English prepositions are hard enough as it is. There are about 100 of them, and they do not translate or match directly into any other language, neither do the other languages between them. So the more prepositional-type languages you learn, the greater the complexity you have to remember, unless you learn them like a child in a small set of chunked phrases and apply those basic patterns across the whole language (that's the easy way but you need access to full phrase audio chunking like we provide here on Glossika training platform).

So instead of some 100 prepositions dispersed with 6 or 7 case endings across 3 grammatical genders like you have in East Europe's Slavic languages, postposition languages give you a total of 12 common endings (or 20 rare ones), across 2 patterns based on vowel harmony. The Finnish-Hungarian-Turkic system is much simpler than the European one.

As you can see at this point, we're not worrying about details, we just want to know what kind of details we will pay attention to in the future.

What I mean by this is if it's a postpositional language, I will pay close attention to those word endings and the kinds of meanings they are telling me about the words they attach to. And that's what master language learners do, they pay attention to those variable patterns and try to guess and elicit meaning from them until it matches the reality they experience and it all clicks on for them (because the next time they hear it, their expectation gets proved to be correct).

Verb Tenses

If the language is an isolating or polysynthetic one (like Chinese, Thai, Vietnamese), this is not a major issue and verb tenses are covered by other words like adverbs of time, duration, frequency, degree -- in the former -- and inserted particles like completion, direction, habit -- in the latter -- but the former also shares many of these traits.

All other languages have some kind of compound verb structure with internal changes to the word, like the English "has been stolen" based on the base "steal".

If you know what elements exist, when you hear the language for the first time you'll be opening your ears to understanding much quicker, which wins over hearing random sounds any day.

The biggest points of confusion for students is the difference between what linguists call tense and aspect.

Tense is just a fancy word for time. That means: future, present, past.

But there are some things that are just factually true, like the earth revolves around the sun. In English we use the present, but that's not completely accurate. A lot of languages will use some non-descript verb, otherwise known as "aorist".

We also express habit in English with the present verb, so it's not good to even call this verb "present". This leads to confusion for both learners of other languages from English, and learners of English from other languages.

The tense for habit is actually "repetition". It happens many times, or regularly. Slavic languages (Russian, Polish, Czech, Croatian, etc) have a specific verb for this for movement. We don't in English. It's simply the present.

Semelfactive is just a fancy word for an immediate one-time occurrence like sneeze, shatter, glance.

Again Slavic languages have a really cool feature where you can add -knu- into the middle of a verb and make it a one-time semelfactive occurrence which also makes it perfective, as explained below.

Aspect is just a fancy word for completion. That means: finished or unfinished.

A finished action is perfective. An unfinished action is imperfective.

Many languages combine Tense and Aspect into one form: past/present finished/unfinished.

A combined tense + aspect finished action is perfect (has done = as of now, had done = as of the past).

A combined tense + aspect unfinished action is imperfect (am doing = as of now, was doing = as of the past).

Individual Conjugations are the Leaves in the Forest

Once you know the BIG META-STRUCTURES of the language, then you can start filling in the leaves.

If you know your language uses either perfective or perfect (probably not both), and you know how to construct and take them from your mental inventory when you want to speak, the final stage is attaching conjugations for each person onto the verb.

Likewise, things like masculine and feminine agreement for nouns and adjectives are also "leaves in the forest" type of details. Most masculine and feminine derive from leftover remnants of a more complicated system like you find in African languages: where nouns have ten classes or more, with specific sounds for each class. This is simplified in Europe mostly with two or three classes, and still based on sound. Classically, -a has been the favored form to describe all things female, so the term "feminine" got adopted to describe this whole class, just like you have a "man" class in Bantu languages which happen to include other things. But most teachers and students on the internet today argue thinking only in the present, as to why certain objects are considered feminine, rather than thinking of the historical development from past to present. Look at the big picture and don't get bogged down with ideology or sensationalism because it makes for a hot topic.

So this is the easiest way to acquire grammar, and it doesn't matter at this point if you get the final step with conjugations wrong. It does matter much more if you use the wrong Tense or Aspect, of course. So get things learned in the right order.

Most European languages follow a very similar verb conjugation pattern. Most Turkic languages have an almost identical verb conjugation pattern. It's one of those things that if you already know it in one, there's very little memorizing or learning to do in a related language.

Final Words

Most of your effort should be spent on Phrase Structure. This means word order and syntax. All of the grammatical details above keep things together.

Your Execution Plan

  1. Download one of our Language Introductions or Word Order ebooks here
  2. Identify the kind of language you're learning and its sounds
  3. Identify its key structure points and the 4 kinds of verbs
  4. Identify the relationship between nouns and verbs (word order and pre/post-positions)
  5. Identify the verbal structure (and use our Word Order publications as a guide)
  6. Start Training with Audio on Glossika with IPA or roman transcription turned on. Do one session every day and don't stop once until you get to 10,000 reps. This is a super basic minimum threshold for familiarity.
  7. Learn the Writing System gradually as you match the sounds with the specific letter combinations and words you've learned: and don't do this before doing audio otherwise it will mislead you
  8. After you've hit 10,000 reps of daily audio practice, start reading grammars to help you fill in the details about anything you'd like to know more about.