Sanskrit Lesson 2 – Dhatu, Magic Roots of Sanskrit

Before reading this second lesson of Sanskrit (Samskrit) it is strongly recommended that you read the first lesson to learn Sanskrit. It is in simple English, easy to understand, interesting and you will learn about a very important unique aspect of the Sanskrit language. In this lesson we will learn about the root words in Sanskrit – Dhatu.

We saw in our first lesson how in Sanskrit we do not give names, but derive names of objects and things based on their properties. Giving a name is just assigning a name that we like to a place, person or thing. Many a times these names are random in naturally evolved languages like English. Deriving a name on the other hand is using a name that tells something about the place, person or thing based on its attributes or properties.

For instance, the name of the place Ayodhya means the one which can never be conquered, derived from Yuddha meaning war. It talks about a kingdom that was never conquered by others in history. Rama means delighting, pleasant, beautiful and Chandra means Moon. Ramachandra hence means as pleasant, delighting and beautiful like Moon.

Unknown Object Identification by Names

Since an object can have multiple properties or features, in Sanskrit same object can have multiple names each describing a property of that object. Note that these are not actually the object names, but names of features of that object.

More than one object can have the same name if they share the same property or feature. By looking at the name of an object in Sanskrit we can guess which object it is without having to memorize its name. In other words, in Sanskrit we understand names not remember them. In case of attributes which are common among many objects, by mentioning a few more attribute names of that object, we can tell which object is exactly being referred to.

Take the case of the word School. If you already don’t know what a School is then you will have to look into the English dictionary for its meaning. On the other hand in Sanskrit, a term used for School is Vidyaalaya, where Vidya means knowledge and aalaya means place. So just by looking at its name in Sanskrit you can say that School is a place where one earns knowledge, or where one learns.

Similarly Shauchaalaya is a place where one can fresh up, Shuchi means clean or fresh. So Shauchalaya means Toilet.

Aushadhaalaya means a medical shop, because Aushadha is medicine, so the place where you get medicine is Aushadhaalaya. Hima means Snow, Himalaya is the abode of Snow, a place where there is snow, The Himalayas. Deva means heavenly, Devalaya is any divine place, like a Temple.

Take the name of the Indian state Meghalaya, in Sanskrit Megha is a term describing clouds. So Meghalaya means Land of Clouds. Meghalaya receives one of the highest amount of rainfall on this planet. Places in Meghalaya like Mausynram, Cheerapunji receive world’s highest rainfall. See how much of general knowledge is hidden in Sanskrit names!

Take the case of the word Bird. In Sanskrit a term used for birds is Khaga, and if you know Sanskrit Grammar, then you don’t need a Sanskrit Language dictionary to know what Khaga means. ga means to move or to go. The English word go is derived from Sanskrit ga. kha means sky. So Khaga is something that moves in the sky – can be used to describe not just birds, but also Sun, even for planes and helicopters! They all move in the sky.

Now see the word Mrga. Mr means kill. So Mriga means the one that moves to kill. All predator animals like Lion, Tiger, etc can be called Mrga. When a person is called Mrga in Sanskrit, it means that person is behaving like a wild animal with predatory instincts, with an intention of harming. Mrgalaya is a name for zoo, a place of wild animals! Cow is not a Mrga, Lion is. Cow is a Pashu. Pashu means being restrained to a specific perimeter. Cows and cattle are restrained by tying them up using ropes.

Tura means quickly. So Turaga means the one that moves quickly. In Sanskrit one of the names of Horse is Turaga. Similarly Ura means belly, uraga is something which moves on its belly. Uraga is used to refer to Snakes, Serpents in Sanskrit.

Dur means difficult, so Durga is something that is difficult to move into or difficult to access. Durga hence is one of the names of Fort in Sanskrit.

In other words, Sanskrit names themselves are like General Knowledge, packed with facts. Just by looking at its names we can tell that a Lotus is pale red in color (Kamala), is born in water (Jalaja), is born in mud (Pankaja), and so on.

If you cannot be sure what an object is by looking at its single attribute name, look for multiple attribute names of that object. One of the reasons why Sanskrit verses use multiple names while referring to the same object or person is so that the reader can be sure which specific object or person is being referred to. Also, as described in the previous lesson, context plays a very important role in understanding the true meaning of a Sanskrit sentence.

So, in Sanskrit, mentioning multiple names act like a filter that further consolidate the object being referred to. If you have one attribute name, and say five objects have that attribute, mentioning any other attribute which other four do not have, will help the reader figure out the actual object being referred to. So usually two or three names are more than enough to make it clear which object or entity is being referred to in a sentence. That is the reason we find multiple names of an entity in many ancient Sanskrit texts.

Of course, as time passed, many entities were referred to by their most common or unique attribute. So nowadays, when a Sanskrit speaker says Kamala, it almost always refers to Lotus. But while reading ancient texts, one has to make sure that the context is actually referring to Lotus and not someone’s reddish face.

No separate Dictionary in Sanskrit – Dictionary is part of Grammar

As we saw earlier, in other languages, say for example in English you just call it Lotus. Now if you don’t know what ‘Lotus’ means in English, then there is absolutely no information you can derive about which object this name represents without looking into an English dictionary.

Even if you are an expert in English grammar, you cannot know what a name means because unlike in Sanskrit, names are independent of the grammar in all other languages on this planet. They are simply categorized as nouns, and you have categories like proper nouns, common nouns – but nothing in the grammar has rules about how to derive a name. In other words, names in other languages are absolute, may or may not say anything about the object, and always refer to a given object.

There is no fixed rule as such in other languages about how you name things. The names are absolute in the sense there is a one-to-one mapping between a name and an object, for instance a Violin is always that, the musical instrument it refers to. Lotus is always that, the flower it refers to. Sometimes you might have multiple objects in English with the same name. For instance, a Mouse might be either an animal, or a computer hardware device. But again, they are absolute names.

On the other hand in Sanskrit, you can use the names to refer to anything that has the attribute being described by that name. For instance, as we saw earlier, Khaga can be used for anything that moves in the sky. You cannot do that in other languages because the names themselves do not describe any properties as such, they are not derived names, but given names.

So while in other languages you require a separate dictionary of names to look into the meaning of words, in Sanskrit all you need to know is Sanskrit grammar and in most cases can easily guess the object from its name. If the name in Sanskrit is referring to a more common attribute, then you need to look into the context of the sentence, or there will be adjacent words hinting at additional attributes of that object with more names, and you can guess the object easily. For instance, if the sentence is about a flower, and says it is pale red and born in water, then it is referring to lotus.

You cannot identify an object in other languages with its name if you do not know the object, even if you are an expert in its grammar. Because grammar has nothing to do with names in other languages. But if you are an expert in Sanskrit grammar, you rarely need a Sanskrit dictionary. In fact, a Sanskrit dictionary similar to an English dictionary is not possible in the first place because objects do not have names in Sanskrit, only attributes do. So even if you write a Sanskrit dictionary, Jalaja should not mean Lotus there, but it should only say,

Jalaja = born in water. For example, Lotus.

And if you know Sanskrit grammar, you will know that Jala is water, Ja is “to be born”. So what is the use of a Sanskrit dictionary then?

Wait, wait. But don’t we need a dictionary to at least say

Jala = Water
Ja = to be born
and so on.

Well as I said earlier, Jala is one of the names of water. Jala in itself is the attribute name that means having a cool touch, which is a property of water. So we can use Jala while referring to water. Thus your dictionary will actually be

Jala = having a cool touch. For example, Water.
Ja = to be born

and so on.

But you don’t need a separate dictionary like this in Sanskrit, if you are an expert in Sanskrit grammar! Why? I will explain, but before that…

Computational Parsing and Structured Information in Sanskrit

In naturally evolved languages like English, sentences can be ambiguous, and this is one of the primary reasons why it is extremely difficult for knowledge representation in Computers using human languages. In the sentence, “Flying planes can be dangerous”, is it the planes that are dangerous, or the act of flying them dangerous?

If Sanskrit were the language of Choice in computation, then you could have directly written compilers to parse Sanskrit sentences, instead of having to invent new programming languages like C or Java. What I mean is, suppose English were well structured like Sanskrit, then you could have written a compiler which directly compiles English sentences into programs, instead of having to invent new syntax for programming languages! The very fact that you have to invent new structured representation for programming languages means that existing grammar is not well structured, is ambiguous and difficult to interpret by computational logic.

If you write a compiler based on Sanskrit grammar, you can have it compile a Sanskrit sentence directly! Of course, the number of keywords in this compiler will be huge, it will be the number of dhatus that I will explain about later.

You cannot do that in other languages. For example, if you had to write a for loop in Sanskrit like how you write in programming languages, you could simply write a Sanskrit sentence which unambiguously says that what computation should be repeated how many times or till what condition is met.

The same holds true for querying stored information. In Sanskrit you wouldn’t need to invent a separate structured database querying syntax like SQL, the Structured Query Language, Sanskrit is already a Structured language and Sanskrit sentences querying information are structured naturally, because the language itself is structured extremely well. If Sanskrit were used then there would be no need for SQL, and database engines like Oracle, MySQL, etc would be just parsing Sanskrit queries, not SQL.

You need SQL today because English is the predominant language in the world which invented computers and computing, and naturally evolved languages like English cannot be used to represent structured queries like SQL because English sentences themselves are not structured well, and are ambiguous. If all those software pundits who invented various computational technology knew Sanskrit, then it would be an all Sanskrit digital world on which Computers would be running today.

In fact, the world’s oldest binary system of representing knowledge using just two symbols is found in the ancient Sanskrit work ChandahShastra by Pingala where enumeration of meters is done using short and long syllables – laghu and guru, similar to how zero and one is used in binary computing.

Many are not aware that Sanskrit is already being used in the very foundation of modern Computer programming languages.

If you don’t know what BNF notation (Backus-Naur Form) is, it is a notation for writing context free grammars and all modern computer programming languages make use of these notations. This idea of writing context free grammar has its roots in the works of the ancient Indian grammarian Panini who used them to describe the structure of Sanskrit words. In fact there are suggestions to rename Backus-Naur Form as Panini–Backus Form!

Parts of Speech – Sanskrit and other languages

If you know English grammar, you must be also aware of the Parts of Speech in English. In the traditional English Grammar you have eight parts of Speech – Noun, Verb, Pronoun, Adjective, Adverb, Preposition, Conjunction, Interjection. Then you have this broad classification of words into Open word classes and Closed word classes, where open word classes include the ones like Nouns, Adjectives etc to which new words can be continuously added as the language evolves. Then you have closed word classes like pronouns, conjunctions etc which are a fixed set of predefined words in English.

Now as we know Noun is the name of a person, place or thing. But there is no grammatical rule in English about how to name a person, place or thing. Similarly there are no rules about how names of verbs are derived and so on. In short, there is no fixed rule about how you can name a word – be it a noun, verb etc. This is the same with languages across the world. I am only using English here as an example. What I say about English in these articles applies to all naturally evolved languages around the world.

So we have two issues here in non-Sanskrit languages. The first is, you will need a separate dictionary independent of grammar, to understand the meaning of different words in English. Grammar and names are totally disconnected in these languages and are independent of each other.

The second natural consequence of this is, the names may or may not give you any information of the object it represents. For instance, while the word Thermometer can imply that it is a device which measures temperature, the word Scissor on the hand implies nothing about what it is!

On the other hand in Sanskrit, a term used to denote Scissor is Kartari, where in Kart means to Cut. So, the term Kartari also tells us what exactly it does, unlike in English.

But we are back to our original question of, how do we know in Sanskrit that Kart means cut, Ja means born, etc?

The answer is that unlike grammar in other languages whose basic building blocks are many different classes called Parts of Speech, the basic building blocks of Sanskrit grammar are just a group of root words called Dhatu.

Dhatu – The magical building block of Sanskrit Grammar

You do not start learning Sanskrit Grammar by learning different parts of Speech, but instead there is an even more fundamental building block called Dhatu. Dhatu is a fixed set of very short words in Sanskrit Grammar representing ideas – any idea like an action, a property, etc. In English they call it Verb Roots, but more specifically these represent ideas like to be, to go, to do, etc. There are 2012 Dhatus in all in Sanskrit, and this is a fixed set.

It is said that ancient Vedic Sanskrit had even more number of dhatus. So the Sanskrit of today, called the Classical Sanskrit, is actually a subset of the original Vedic Sanskrit and that is for only one reason, the number of dhatus was reduced in post vedic period.

Everything else in Sanskrit Language is built on top of these 2012 Dhatus. If you know the meanings of these Dhatus, you can derive the  meaning of ANY Sanskrit word! That is because all Sanskrit words are built on top of these Dhatus. Each word is derived from one or more Dhatus using the rules of Sanskrit grammar. So Sanskrit never needs any loan words, because the very process of word creation is inbuilt in Sanskrit grammar.

Unlike non-Sanskrit languages where Dictionary and Grammar are independent of each other, Sanskrit starts with a dictionary of Dhatus and Sanskrit grammar is just the rule of creating words and forming sentences using words derived from these Dhatus!

You should by now have understood what I meant when I said you don’t need a Sanskrit dictionary if you are an expert in Sanskrit Grammar. If you know Sanskrit Grammar, then you also know the Dhatus which are the basic building blocks of Sanskrit, and if you know them you also know the meaning of every word, because all Dhatus have meanings and all words in Sanskrit are derived from these Dhatus. So you will never need a separate dictionary to find meanings of names, because names themselves are meanings in Sanskrit.

If you are from a computer programming background, then Dhatu words are like base classes, and all other words in Sanskrit are like derived classes. They represent various attributes, and when you apply these attributes to specific objects, they become like instances of those classes. For instance, Mr and Ga are base classes from which the class Mriga is derived, which means anything that moves to kill. Now when you apply this attribute to a specific object like say a Lion, it becomes an instance of this derived class Mrigam.

More on this instance creation later. For now just remember that Dhatu is a abstract base class, vyaya is a derived class and avyaya are instances of derived classes. Dhatu is abstract because you don’t create instances of abstract classes, you derive Vyaya words from Dhatu, and then create instances of those Vyaya words i.e Avyaya words. There will be a separate detailed lesson on this later. So don’t worry much if you don’t understand this yet.

Samskrit – A Refined Language

Sanskrit has remained a language unchanged, never evolved but was perfectly designed in the very beginning with everything in place. No new grammar rules were added to Sanskrit at any point of time later. All new words created in Sanskrit can be traced back to a combination of these 2012 dhatus and related grammar rules, and also retaining the original idea of those dhatus. So you don’t need ever expanding dictionaries in Sanskrit as new words are created, because they can easily be split into their root dhatus to extract the meanings of these new words.

In Sanskrit the set of Dhatus remains fixed, and all new words are derived from these Dhatus. But dictionaries of other languages keep increasing over time, because there words are independent from grammar. So for instance, the English dictionary is ever expanding, started with around 3000 words, and today has nearly 300,000 words!

For most of these words you need to have a dictionary of English to find its meaning, where as in Sanskrit you can create millions of words and still there wouldn’t be need for a dictionary! Just split the words into its Dhatus and you will get the meaning! And Dhatu is a closed word class in Sanskrit grammar, meaning you cannot add new dhatus to the list.

In fact, in Sanskrit, if you are creative enough, you can write your own dictionary that also doubles up as a knowledge base describing huge number of facts about different objects and classes.

Those who known modern English find it next to impossible to read and understand old English, or for that matter those who know modern Kannada (Hosagannada) cannot understand Old Kannada (Halegannada), same in other languages as well. But in Sanskrit, there is nothing like modern, old etc because there has been no evolution of Sanskrit in the first place.

EDIT: The 3 paragraphs below were added after reading the comments by Dharma Dhwaja in the comments section.

The only change was reduction in the number of dhatus from Vedic Sanskrit to Classical Sanskrit. This happened after Panini wrote Ashtadhyayi that hugely simplified the original grammar and by removing dhatu forms from the original Vedic Sanskrit whose usage had become rare by his time. So, if the Vedic Sanskrit was like the difficult C programming language, Panini created a Java version of it which became the Classical Sanskrit.

So, it was not that Panini created the grammar of original Sanskrit, the Sanskrit was designed even before the vedas were written, because they are written in Vedic Sanskrit and you need to have the language first. Panini simplified the Vedic Sanskrit, and it became Classical Sanskrit.

In other words, Sanskrit never evolved. In one shot, in the beginning, Vedic Sanskrit was designed. Much later, in one shot, a simplified version of Sanskrit, a version 2.0 – Classical Sanskrit was created by Panini that became quite popular in writing later Sanskrit texts, because it was more easy than Classical Sanskrit.

In terms of ease of use – if Vedic Sanskrit is like the C programming language, then Classical Sanskrit is like C++.

An introduction to some more Dhatus

The entire process of learning Sanskrit is learning Dhatus and the rules of playing around with these Dhatus creating extremely beautiful and innovative combination of words and sentences. There is no unnecessary complication. We will have a very brief look at some Dhatus now, and as we move forward in future lessons, make ourselves more comfortable with more Dhatus and the rules of using Dhatu to form words and use them in sentences. As I said in the beginning of this series in the first lesson, this Sanskrit learning series will be more like practical classes, than plain boring theory classes.

We now know that Dhatu is a basic building block of Sanskrit words. All other names in Sanskrit are derived from these fixed set of Dhatus. When we said earlier that Khaga denoted a bird, implying the one that moves in the sky, we saw that this meaning came from splitting the word in kha and ga where kha meant sky and ga (from the dhatu gam) meant to move. So by now it should be clear that in Sanskrit to understand the meaning of a word, all we need to do is split it into its root Dhatus and using the meaning of the ideas behind that dhatu we can understand the meaning of the word. So simple and beautiful, isn’t it?

Dhatu Roopa – Splitting words into Dhatus

This processing of splitting a word into its dhatu format is called Dhatu Roopa. Remember this term, as we will be using it quite often. Dhatu Rupa means the Dhatu Form. By Dhatu Roopa we mean finding out the root Dhatus of the word, i.e doing the reverse process of word creation using Dhatus to find word meanings.

Let us start with the very word Dhatu, because even this is a Sanskrit word and hence should be derived from some Dhatu word This word is derived from the Dhatu called Dha in Sanskrit. Dha means foundation, root, basic building block. How is the word Dhatu derived from Dha? More on this in future lessons. For now, just remember that Dhatu is derived from the Dhatu word Dha.

Since, the meaning of this is root or foundation, all the root words of Sanskrit which form the building block of Sanskrit language are called Dhatu. Moreover, as we saw earlier, since these are names of the properties, and since the property name Dhatu represents root, foundation, basic building block, it can be used to represent any such root or base object!

So in Chemistry for instance Dhatu represents Chemical Elements, Metals etc which are the basic building blocks there. In Ayurveda, Dhatu represents the basic building blocks of our body like for instance Asti Dhatu represents the building blocks of bones, as Asti represents Bone in Sanskrit. Rakta Dhatu represents the building blocks in blood, where Rakta represents Red Color and hence Blood in Sanskrit.

Kr is a Dhatu which means to do. Karman is a derived word of this Dhatu meaning deed. Kriya is derived from this dhatu and means action. The word Prakriya is derived from this dhatu and means process. Then the word Sakriya is derived from this Dhatu and means being active. And so on. In fact there is a huge number of combinations possible from each dhatu, and we will learn about the actual process of creating words, combination of words, sentences, meanings and so on in the future lessons of Sanskrit grammar.

Inflection – an amazing contribution of Sanskrit to Linguistics

But note that, one of the biggest contribution of Sanskrit to the world of linguistics was inflection. Consider the English word create – its inflections are words like creating, created, creation, creates, creator, etc. Sanskrit was the first language in the world to come up with this concept of using the same word, modify it a little (inflect it) and use it to mean things related to that word.

This is a great innovation, which many of us ignore, just like the way we ignore the wonderful idea of place value based number representation invented by ancient Indians.

Imagine having to create a separate word for created, creation, creating, creates, creator etc without using the inflected forms! Sanskrit gifted the concept of inflections to the world of languages.

Summary of Sanskrit Lesson 2

Today we learnt that

• In Sanskrit, attributes and properties have names, and all the names in Sanskrit are derived from a fixed set of 2012 root words called Dhatu.
• Dhatu, not the Parts of Speech, forms the basic building block of Sanskrit.
• The process of deriving names is in built in Sanskrit Grammar, because of which Sanskrit never requires any loan words from other languages. If there is a new invention, a new object or a new information discovered, Sanskrit grammar can be used to easily create one or more new words to represent it. We saw an example of representing download and upload in our First Sanskrit Lesson.
• Since the Dhatus have meanings attributed to them, and since there is a predefined process of deriving names in Sanskrit, all names in Sanskrit have meaning inherent in the name itself unlike in other languages. For example in English the word Quiz means nothing without a dictionary, or the word Magma means nothing without a dictionary. However in Sanskrit, every word means something on its own, without referring to an particular object or class. In other words, all Sanskrit names state facts – describe the nature and attributes of the thing they represent.
• Since Sanskrit is an extremely well structured language with no ambiguity in its grammar , Sanskrit Sentences can easily be used in computational language unlike other natural languages whose sentences are ambiguous and whose grammar is extremely complex making it difficult to write compilers which can understand English sentences. For instance, if Sanskrit was used as a language for database queries, you wouldn’t have needed SQL, because queries in Sanskrit are as structured as SQL.
• Dhatu words have meanings over a vast range covering all possible basic meanings representing all human knowledge and actions. Words are derived from one or more Dhatus using a set of grammar rules to represent compound properties and attributes like we saw for “moving in sky”, “born in water” and so on. These attributes are then used to represent objects which have the properties matching these attributes, as we saw for Birds, Lotus, etc.
• So Sanskrit language words are an encyclopedia in itself, with each name describing one or more properties of what it represents.
• More in next lesson. Questions, corrections, criticism is welcome. Please do not forget to share this lesson. Knowledge and Happiness grows by sharing.

For your reference, a Complete List of Sanskrit Dhatus

Continue to Sanskrit Lesson 3 – The Science behind the amazing Sanskrit alphabet

Download HitXP Mobile App