Unit III

Linguistics
3.1 Introduction to Linguistics
Importance of Language

• Communication:

o Fundamental for interaction (e.g., asking "how are you?", "did you have
breakfast?").

o Crucial for scientific discovery, advancement of knowledge, and collaborative

work.

• Trade, Science, Technology & Societal Progress:

o Communication underpins progress in these areas.

o Effective language processing is vital.

• Artificial Intelligence (AI) and Natural Language Processing (NLP):

o Language capabilities are central to AI and NLP applications.

o Systematic study of languages is essential for advancing science and

technology.

Relevance of Ancient Indian Contributions

• Understanding ancient contributions can provide insights into language.

• Studying language structure helps:

o Preserve the underlying principles that govern a language.

o Ensure that wisdom from ancestors is not lost.

Components of Language

• Skills:

o Receptive Skills: Receiving language (e.g., Listening, Reading).

o Productive Skills: Producing language (e.g., Speaking, Writing).

• Medium:

o Sound

o Script.

Putting It Together: Skills & Medium

• Sound as medium:
 o Receptive Skill: Listening.

o Productive Skill: Speaking.

• Script as medium:

o Receptive Skill: Reading.

o Productive Skill: Writing.

• These form the four fundamental components of a language.

Requirements of a Language

• Languages need robust structures and mechanisms to facilitate effective Listening,

Speaking, Reading, and Writing.

Linguistics

• Branch of language research.

• Scientific study of language.

• Systematic study of language to understand:

o Speech.
o Sounds.

o Grammatical structures.

o Meaning.

• This understanding enables all receptive and productive skills.

• Helps analyze language form and meaning.

• Identifies systematic methods to derive word forms and their meaning.

• Uses structured rules and syntax.

3.2 Aṣṭādhyāyī

Introduction
 • This section delves into how syntax is developed and how to make sense of form and
sound in language.

• The earliest known approach to this is Panini's work, a grammarian from the 6th
Century BCE (approximately 2800 years ago).

Panini and Aşțādhyāyi

• Panini's work serves as a great way to understand the Indian contribution to

linguistics.

• Aşțādhyāyi:

o Created by Panini.

o "Ashta" means eight, and "adhyaya" means chapter.

o The work consists of eight chapters.

• Panini's work is the culmination of a long grammatical tradition.

Panini's Approach

• Panini did not invent a language from scratch; he reverse-engineered it for easy
understanding and logical approach to Sanskrit.

• He observed the existing language use and patterns to frame proper rules of grammer.

• 3983 Rules (Sutras):

o Panini wrote 3983 rules, known as Sutras.

o These rules were designed to accommodate the patterns and variations in the
Sanskrit language.

o He fitted these patterns using the rules.

Structure of Aşțādhyāyi

• The 3983 rules are arranged into 8 chapters.

• Each chapter is further divided into 4 quarters.

• 8 chapters x 4 quarters = 32 sections.

Samskritam (Sanskrit)

• Panini's work (Ashtadhyayi) refines and structures the language.

• Samskritam means "refined."

• Sanskrit is considered a very refined language.

• The Ashtadhyayi is viewed as the best available descriptive model of a language.

Later Commentaries
 • Katyayana's Varttika:

o Composed about 200-250 years after Panini.

o Varttika means commentary.

o Commentary on Panini's work.

o Addressed loose ends and clarified Panini's work.

o Provided additional notes.

• Patanjali's Mahabhashya:

o Commentary on the Ashtadhyayi.

o Created in the 2nd Century BCE (another 200 years later).

o "Bhashya" means commentary, and "Mahabhashya" means great commentary.

o Katyayana's Varttika and Patanjali's Mahabhashya, together with Ashtadhyayi,

provide a basis for understanding the science behind the Sanskrit language.

Distinguishing Aspects of Panini's Grammar

• Complete Vocabulary:
o The entire vocabulary of the Sanskrit language can be generated using the
3983 rules (approximately 99.9%).
• Handling Exceptions:

o Special handling for the few exceptions.

• Sutras (Aphorisms):

o The rules are known as Sutras (aphorisms).

o Easy to memorize.

o Concise.

o Provide crisp rules.

• Mastery of Sanskrit:
o Familiarity with the rules leads to an unambiguous mastery of the Sanskrit
language.
o One effectively becomes a "walking Sanskrit language" without needing
dictionaries or thesauruses, once a person masters these sutras or rules.

Educational System and Sanskrit Mastery

• The traditional Indian education system (until the 19th century CE) ensured students
mastered this language.
 • Introduction of the Macaulay system of education resulted in a discontinuity.

Rule-Based Processing and Word Generation

• Language processing and word generation in Sanskrit are strictly rule-based and
derivative.

• Similar to a mathematical equation: entire words can be derived through rules.

• Computer-Driven Methodology:

o Suitable for computer-driven methodologies.

o Few additional assumptions are required.

Distinguishing Aspects of Panini

• Word derivation uses a step-by-step process.

• Modular approach to word generation.

Word Formation

• Involves combining basic components:

o Verbal root or nominal stem.

o Adding suffixes.

o Invoking rules.

o The process results in a valid word.

• Unique, algorithmic approach.

• Vocabulary is not fixed.

• New words can be generated as long as the rules are not violated.

Panini's Work: Data Structures and Computational Elements

• Panini deployed interesting data structures and computational elements.

• This makes his work unique among linguistic studies.

Vedanga and Panini

• Vyakarana or grammer is one of the Vedangas, or the limbs to understand the Vedas.

• Panini's work puts together grammatical ideas.

3.3 Phonetics

Introduction

• Phonetics is the starting point for language, especially for cultures with a strong oral
tradition.

Oral Tradition and the Vedas

• The Vedas, transmitted orally for thousands of years, have survived due to scientific
methods of oral rendering.

• The preservation of pronunciation indicates the development of a science of

phonetics.

• Good phonetics ensured impeccable textual transmission, surpassing classical texts of

other cultures.

UNESCO Recognition

• Vedic scriptures, transmitted orally, are unique.

• UNESCO recognized the Vedas as a Heritage for preservation in the form of Oral
Knowledge.

Phonetics in Indian Knowledge Tradition

• The study of sounds in a language, particularly sound production and communication,
is part of phonetics.

• In the Indian knowledge tradition, the science of sound study is known as Shiksha,
one of the six Vedangas.

Pratishakhyas
• The IKS Corpus includes Pratishakhyas, which address the issue of how sounds are
produced.

• Rigveda Pratishakhya and Taittiriya Pratishakhya are the earliest works on the
subject.

• They lay down the rules of phonetics.

Panini and Phonetics

• Panini discussed phonetics in Ashtadhyayi through sutras.

• He also created a separate addendum known as Paniniya Shiksha.

• According to Panini, the origin of sounds is specific to the letters (varnas) in Sanskrit.

Origin of Sound in Oral Cavity

 • Six locations in the oral cavity are identified as sound origins.
• Sounds originate from these locations or a combination thereof.

• These areas include the nasal area, palate, lips, and throat.

• All of these are involved in the production of various varnas.

Example: Kantha (Throat)

• The Kantha (throat) generates sounds like "a", "k", "kh", "g", "gh", "ha".

Paniniya Shiksha
 • It details how to pronounce sounds and their origins.

• Includes aspects like nasal sounds (m, n, n).

• Recognizes articulatory effort: throat, nose, palate etc.

Vowel Variations
• According to Panini, vowels (a, i, u, e, o) have 18 possible variations categorized as:

o Hrasva: Short

o Dirgha: Long

o Pluta: Extended (held for 1, 2, or 3 counts)

Svaras

• Three types of Svaras(tones) are specified:

o Udatta: Raised Tone

o Anudatta: Unraised Tone
o Svarita: A combination of the two

• There are three possible variations:

o Same level (sound from the earlier variation is the same level).

o Raised(sound from the earlier variation is braced).

o Lowered(sound from the earlier variation is braced).

Nasal and Non-Nasal Sounds

• He talks about Nasal and Non-nasal sounds:

o a is nasal

o a, i is not nasal

o i, ii, is nasal you have to use your nose also in making the sound i

Alpaprana and Mahaprana

• Consonants are categorized as:

o Alpaprana: Little air (e.g., k).

o Mahaprana: More air (e.g., kh) *When you say the th, if you put your hand
in front of your tongue you will find a lot of air will blow. If you say k no air
will blow

Efforts in Analysis
 • Identifying the place of pronunciation and the required articulatory effort involved a
lot of effort, observation, and analysis.

• Panini took all of this into consideration.

• The recitation of Vedas is the gold standard for phonetics and preservation.

Benefits of Good Phonetics

• Accrued benefits from good phonetics in a language.

• Possible to impart good phonetical training to language aspirants.

• Closely monitor and rectify phonetical inaccuracies.

• Arrest deterioration of pronunciation over time.

• Ensures preservation and transmission of language components through oral tradition.

Summary

• The greatest contribution of the Indian Knowledge System in linguistics is through

Shiksha.

• Panini accommodated all of these elements into his grammatical work, Ashtadhyayi.

3.4 Word generation

Introduction

• After discussing phonetics, we now examine Sanskrit word generation as created by

Panini through his Ashtadhyayi.
• The ultimate building block of any language is the word.

Language as Assemblage of Words

• Language is an assemblage of words.

• Words are combined in various ways to communicate ideas and transact knowledge.
 • Understanding word formation and generation is central to language mastery.

Sanskrit Grammar and Word Generation

• Sanskrit grammar addresses word generation in a unique way.

Example: Verbal Root 'kru' (to do)

• Let's use 'kru' (to do) as an illustrative example.

• 'Dukrn': Panini attaches indicators to the front and back of verbal roots for
mathematical transformation purposes.
• In Panini's Dhatupatha (list of verbal roots), the root is represented as 'Dukrn'.

• By removing 'du' and '[FL]', we get the substantive part of the verb, 'kr'.

• 'Kru' means something to do with doing.

• Example: Karma comes from kru.

Word Derivatives of 'kru'

• Karoti: Does (present simple tense).

• Kurvan: Doing (continuous).

• Karta: Doer.

• Krtva: Having done.

• Karotu: Please do.

• Kartavyam: It must be done.

• Kartum: To do.

• Krtam: Done (past tense).

Other Verbal Roots: 'pat' (to read) and 'gam' (to go)
• We see two more verbs: 'pat' (meaning reading) and 'gam' (gacch related to going).

Similar Word Patterns

• The equivalent words for the two roots follows a similar pattern like for kru:

o Karoti: Patathi, Gacchati

o Kurvan: Pathan, Gacchan

o Kartha: Pathita, Ganta

o Krtva: Pathitva, Gatva

o Karotu: Pathtu, Gacchatu

 o Kartavyam: Pathtavyam, Gantavyam

o Kartum: Pathitum, Gantum

o Krtam: Pathitam, Gatam

• If you know the process for a verb, you can apply it to approximately 2200 verbs.
Word Pattern Formation Logic

• There is a consistent logic for creating patterns of words in Sanskrit.

Base and Suffix

• "kr", "path", "gach" are called Base.

• In Sanskrit, one start with the Base.

• The basic building block in Sanskrit, you start with the Base, then add a suffix to it
and then get a word, but in the process of getting the word, some more rules will be
triggered

Suffix Example

• Take the verb 'kr', then add the suffix 'tu' > "karotu". The Suffix "tu" triggers karotu.

All Possibilities in Word Generation

• To make rich combination generation of words in Sanskrit Language, a few things

need to be kept in mind.

• Start with a verbal root.

• Add suffixes.
• Add more suffixes, leading to a verb form and finally a complete word.

Noun Root and Suffix

• Starting with a noun root, you add suffixes.

• Some suffixes can give you masculine, and some suffixes can give you feminine.

• Start from Noun Root and add Suffix and get Noun Form.

Verb to Noun Conversion

• Add a suffix to a verb to convert it into a noun (e.g., "Do" -> "Doer").
• Convert noun to verb is also possible.

Example: Nominal Root 'Ram'

• Take Nominal Root called Ram.

• 7 x 3 = 21 suffixes are available.

 • Using these, 21 different word forms can be created.

• For example adding 'su' to "Ram" gives you "Ramah".

Combine Root with Suffix with rules

• Add any of the suffixes to a nominal root to get a valid noun form.
• Rules will govern the combination.

• Take a verbal root like 'path' and add suffixes like 'tip'.

Understand the Logic

• Take a nominal root and add suffixes, resulting in valid nouns.

• Take a verbal root and add suffixes, resulting in a verb form.

The Numbers: Cases, Singular, Dual, and Plural

• 7 cases and singular, dual, and plural (3), this defines the number of suffix options.
• First, second, and third person and singular, dual, and plural (3 x 3 = 9) defines the
additional types of suffix combinations.

Word Generation Logic

• The word generation is algorithmic in nature.

• Take something (the root), add something on top of it (suffixes), apply some rules,
and derive the final word.

• Rule-based mechanism which will be discussed again when looking at computational

elements of Ashtadhyayi.

3.5 Computational aspects

Introduction

• The mechanism in Ashtadhyayi is: take a word (a base), add a suffix, and apply rules.

• This methodology is similar to many computational concepts.

Panini's Computational Concepts

• Examine the computational concepts that Panini used in Ashtadhyayi approximately

2800 years ago.

Computer Languages vs. Panini

• Computer languages are formal with their own vocabulary and syntax.
• Instructions are given as algorithms or programs.
 • Panini's grammar shares many of these characteristics.

• Similar ideas of today could be found in Panini's grammar, because that is how whole
Sanskrit language is sort of put together

Paninian Approach: Mapping to Computational Concepts

• The Paninian approach to Linguistics and Sanskrit grammar maps to certain

computational concepts.

List of Computational Features

• Exclusive Syntax:

o Panini uses an exclusive syntax for Ashtadhyayi.

o Certain conventions must be followed for Ashtadhyayi to make sense.

• Exclusive Vocabulary:

o A vocabulary is specifically meant for Panini's work.

o Terms like 'tip' and 'mat' have specific meanings.

• Abbreviated Forms and Mnemonics:

o Abbreviated forms and mnemonics are used for brevity and better retention of
ideas.

• Algorithmic Approach to Word Generation:

o Takes a strictly algorithmic approach to word generation.

o Sometimes uses recursive logic.

Amenability for Machine Coding and NLP

• These aspects make Sanskrit easily amenable for machine coding and current-day
applications such as Natural Language Processing (NLP).

• The greater these features are in a language, the greater the possibility of using the
language structure, syntax, and grammatical foundations for NLP.
Sanskrit as an Attractive Option

• Sanskrit offers a very attractive option because it has all these features. Therefore,
sanskrit as a language to be studied and spoken holds immense importance.
3.6 Mnemonics

Introduction

• This section begins with the basic building block of Panini's entire scheme of things.

• These are called Maheshwara-sutras.

Maheshwara-sutras

• Panini's entire Sanskrit grammar rests on these 14 fundamental sets of sutras known
as Maheshwara-sutras.
• These are numbered 1 to 14:

o ai u N

o r1K

o eoN

o ai au C

o ha ya va ra T

o la N
o na ma na na na M

o jha bha N

o gha dha dha S

o ga ba ga da da S

o kha pha cha tha tha ca ta ta V

o ka pa Y

o sa sa sa R
o ha L

• He has used this fundamental set of sutras to refer all grammatical things.

Observations about the Sutras

• Each Sutra is an ordered set of letters (e.g., ha ya va ra T).

• Set 1 is a, i, u

• The consonants at the end of each (like N) is to be removed and have to be ignored.

• a i u N means a i u.

• g ba ga da da S; S has to be removed.
 • Sutras 1-4: Vowels (a, i, u, r, l, e, o, ai, au). The color code is yellow.

• Sutras 5-14: Consonants.

• Example: ha L ends with l.

Termination Tag
• First Sutra a i u N ends with N

• 13th Sutra sa sa sa R ends with R

• These (N, R, etc.) are considered merely as a termination tag.

• They are not included in calculations.

Distribution of Letters

• ka Varga (ka, kha, ga, gha, na).

• These are distributed throughout the sutras.

o Sutra 7 has na.
o Sutra 9 has gha.

o Sutra 10 has ga.

o Sutra 11 has kha.

o Sutra 12 has ka.

• Other groups are also scattered.

Intelligence Behind the Arrangement

• The order appears obscure.

• However, understanding the transformation rules reveals the intelligence behind it.

• Efficient rules in an arbitrary fashion is the charm of Panini's work

Panini's Use of Mnemonics

• Panini used plenty of mnemonics.

• Examples:

o a i u N (combining a, i, u and N).

o ka pa Y.
o ga ba ga da da S.

Operation of Mnemonics

• If you say ac:

 o Start from 'a' and continue to 'c'.

o The ordered set is all letters in between

• This is the formula he uses.

• Suppose you say ha L it means whole set.

• So ha L means consonants set.

• Suppose khay; he will use all that in his 3983 rules in several places he will use.

• The ordered set is kha, pa, ca, ta, ca, ta, ta, ka, pa. You have to look where kha is and
keep on taking all up to where i is

• Then that represents the list column 1 and 2 ka, kha, ga, gha, na, ca, ch, ja, jha, na
etcetera.

Data Structures

• These are kind of data structures, data representation and things of that kind through
the mnemonics.

Sutra Understanding

• Example Sutra: iko yan-aci.

• This contains three mnemonics:

o ik.

o yan.

o ac.
• Basically what he is saying if ik is followed by ac, then simply replace ik by yan.

• If one knows mnemonics as per Panini, one can show a complex idea very beautifully.

• Firstly, said that Ik is followed by ac.

• In Maheshwara-sutra, it means starts from i goes up to k:

o You have to ignore i,r.

o Ik = i,r.

o ac; known that a i, u, r, l, e, o, ai, au.

• yan; 5th Sutra is ha ya va ra T 6th Sutra la N > y, v, r, 1

• There are 3 sets A, B, C

• A has i, u, r, 1

• B is consists all of vowels

 • C contains y, v, r, l

Application of Sets

• These sets share numbers, meaning you are always dealing with the same number of
entries

• Already set the rule.

• Whenever, same set, apply set correspondence:

o Replace i to y.
o Replace u to v.

o Replace r to r,r.

o Replace 1 to 1.

• It becomes a simple idea, how he constructs mnemonic.

Example: iko yan-aci

• Consider the words "prati" and "ekam”.

• "prati" = prat + i
• "ekam" = +e

• Part of ik that is i

• e is part of ac

Rule Application

• If "ik" is followed by "ac", then replace "ik" with "yan".

• 1-1 correspondence.

• Replace i with y
• Remove i and instead e, prat + e+y + kam (Pratyekam)

• Understands step by step, but still if you know ik, if you know is yan, if you know if
know is ac then can do complex operations
Efficacy of Mnemonics

• If you understand the power of mnemonics, one may be able to construct any sentence
in sanskrit out of will.

• Richness of using mnemonics a is the special charm of this part of the data structure.

Summary
 • Mnemonics are very rich and unique data representations to give instructions in
different ways.

• Panini uses mnemonics very intelligently to construct rules.

• Panini's rules around mnemonics makes it easy to remember.

3.7 Recursive operations -Introduction to use of Kaprekar Constant 6174 in recursion

Introduction
• In this section we examine another aspect of Paninian grammatical operations.

• Ashtadhyayi employs Recursive Logic to process grammatical applications.

• One example: the logic of Samasa (a method of creating compound words).

Samasa

• Samasa means creating compound words from a group of nouns.

• Like in other languages, Sanskrit can combine one or more noun forms into a single
noun (a compound word).

• The process of creating a single noun from several nouns is called Samasa.

Recursive Logic in Samasa

• Seeing the Samasa principle as an example below.

• This is how Panini applies recursive logic to obtain compound words.

Example: "shastra" and "nipuna"

• Consider two words:

o shastra : in shastra (scripture)

o nipuna: expert

• Let us say we want to put these two words into two Nouns into a single Noun.

• Purva-pada: First word (shastra).

• Uttara-pada: Second word (nipuna).

Root, Suffix, Word

• In Sanskrit, noun forms are generated from a noun root using suffixes.

• Example: Root, Ram + suffix > word Rama .

Dismantling Suffixes

• Suffixes of the two words can be dismantled to obtain the respective noun roots.
 • Dismantle suffixes to get the noun roots.

Apply Suffixes back to root

• Remove the Suffixes, and the root can become a real word again

Combining roots
• The compound word can be obtained by combining the purva-pada and the uttara-
pada.

• shastra + nipuna = shastra-nipuna.

• shastra-nipuna becomes a new noun root.

Generating Different Forms

• Apply suffixes to this new noun root.

• Get all the different forms (as seen earlier).

• You can do it any number of word.

Generalizing the Process

• Let W be the set of noun roots to combine.

• Remove suffixes to get noun roots (w1, w2, w3... wn).

Recursive Compound Word Formation

• Compound word can be very recursively done.

• At stage “n”, there already be a compound word from the previous stage > Sn-1 .

Adding the Nth root

• Simply add the noun root of the n th stage to the compound root at the nth stage.

• Sn = Sn-1+ wn
Recursive Logic

• Read inputs n and W = w1 + w2 :Set of noun roots to be combined.

• Let the first noun root be the purva-pada and let the compound word be the purva-
pada >a single word =a compound word.

• For i = 2 to n

• Uttara-pada = wi (take the i =2 >take second one =w2 .

• Simply add it to the previous word (Sn-1) >you get the new word at stage I stage
(Stage2).
 • Now replace purva-pada + the new compound word +take the next word uttara-pada
and one can go like that until exhaust the word

Illustration with five Nouns

• Recursive logic = Sn= w1 +w2+w3 ... wn

• The noun root for the compound word.

• Attach suffixes.

• What structure is what is this structure?

• Possible that is what is the structure

• dipa, placed, sthithasya :dipa placed in belly .udara

• Ghata,

• of pot, pierced =nanachidrasya

Combining the root with suffixes

• Remove all the Noun to generate the roots.

• To generate nouns which sort of combined :nanachidra + udara+ sthitha +dipa

Applying recursive Algorithm

• Use the recursive algo=n=5

• W, letters that you have.

• PP = nanachidra

• purva-pada ,

• 1st compound,

Implementing recursive Algo

• i=2

• i=3

• i=4

• i =5 =the final noun root= nanachidra, ghata, ghat =udara sthitha, dipa.

• Attach now different cases ( 1, 2, 3 ) and do what needs to be done.

3.8 Rule based operations

Introduction

• Examine another aspect of Panini's computational elements.

Rule-Based Processing
• Sanskrit language uses rule-based processing.

• Panini's system applies grammatical conditions to derive words (rule-based engine).

Structure of Rule Application

• A situation is presented.

• Sutras provide the necessary and sufficient conditions to address the situation.

• Basic logic in deriving any word in Sanskrit.

• Sutras resemble the structure: If this condition is satisfied, then perform this
operation.

• We saw "iko yan-aci" in mnemonics.

Computer Program Analogy

• The example in mnemonics is like: if this condition is satisfied, do this.

• All sutras are similar to computer instruction programs.

Derivational Grammar

• Sanskrit grammar is entirely derivational.

• Every word can be derived from the start.

• Every conceivable word can be derived using a set of rules in a strict algorithmic
fashion.

The Process: Rule Search and Application

• Start from a Base (Verbal Root or Nominal Root).

• The logic can search if any of the 3983 rules can be applied.
• The root goes through some transformation.

• At every stage, inquire if any of the 3983 rules qualify to be applied.

• If a rule qualifies, apply it, and move one step further in the transformation.

• The procedure stops when no more rules can be applied, resulting in a valid final
word.

Amenability to Computer Processing

 • This rule-based recursive structure is highly amenable to computer-based processing.

• Similar to how a computer takes instructions and uses IF and THEN conditions to
derive results.

Flowchart Representation

• Start

• Read Inputs:

o Let i = 0 (initial rule counter).

o i = i + 1 (take the first rule).

• First, check whether end of rules is reached?

o If No, read the Sutra for the ith rule .

o Is Sutra applicable?

o If Yes, Perform operation.

o Apply Transformation; reset counter to 0.

o If No, increment i = i + 1; go to the next sutra.

• End of 3983 Rules?

• To find the final result by quitting and by print with logic.

• The end result being a valid word.

• (Efficient algorithms avoid scanning all 3983 rules every time.)

Example: Deriving Rama (By Rama)

• Take the noun root Rama and derive the singular form of the third case (which means
"by Rama").

• Step 1: Noun root Rama is taken.

• Scanning the set of rules, in 4th chapter, 1st section, 2nd rule of Ashtadhyayi apply to
get the first applicable rule.

• 4.1.2 supplies a suffix called “ta” for generating the third case.

• Therefore, we have Rama + ta.

Applying Further Rules

• Scan again and find the applicable rule.

• 7th chapter, 1st section, 12th sutra becomes eligible.

• That rule will be applied.

 • It provides conditions for replacing ta with something.

• Applies in this case and replaces “ta” with ina, as the rule suggests.

• Now, we have Rama + ina.

• Scan the rules again.

• 8th chapter, 4th section, 2nd rule becomes applicable.

• This is actually guna sandhi.

• Since a is "followed" by i, 6th chapter, 1st section, 87th apply and we find it will be

o “a” + “i" will transform to "e" based a.

• As per the rule, the vowel a > a + will transform to "e" per guna and if it's apply " it
will become Rama +n.

• Scan rules.

• Chapter 8 -Section 4 -Rule 2 .

• n will be n ""

o Rama + n will become Rama + n.

o And that's the stage you have now .

• The rules will not be applicable ,so the procedure will stops !

• 3rd case Rama is ""Ramena!""

• What a charm to transform by "a applicable rule""

• The transformations go until we find "a valid word"""

• Moment a transformation happen you transform again through scanning of rules

• This is the charm of Panini's Base Rule Operations!

• If we remember the application the valid output will be derived.

3.9 Sentence formation

Introduction

• Examine the logic for Sentence Construction in Panini's grammar.

• There is a logic or method in constructing a sentence.

Basic Logic of Sentence Structure

• Verb Requirement:
 o A sentence must have a verb (explicit or implicit).

o Verb denotes an action.

o Without an action, there is no sentence.

o Example: "Dosa" alone is meaningless without a verb.

• Association of Verb with Other Words:

o Verbs must be associated with other words.

o These other words denote the participant and other attributes connected to the
action.

o Every language has this feature.

• Comprehensiveness:

o merely uttering Come/ Leaves has many ambiguities.

o uttering Ram comes or Yussuf comes with his friend is more comprehensible.

Basic Building Blocks of a Sentence

• The verb is fundamental to the sentence.

• Verbs cannot hang alone; they require other parts of the sentence.

Sanskrit Sentence Structure

• Examine how Sanskrit and Indian languages structure sentences.

• Compare with English.

English Sentence Example

• “The fat boy eats the tasty food with the hand”.

• Sanskrit Translation:
o sthula balaka svadu bhojanam hastena khadati

o sthula balaka = a fat boy

o svadu bhojanam = tasty food

o hastena = with the hand

o khadati = eats

Word Order Flexibility in Sanskrit

• Jumbling the word order in English leads to nonsense.

• Jumbling the word order in the Sanskrit sentence does not change the meaning.
 • Reason:

o Sanskrit cases are attached to the verb.

o The adjective can be identified by the cases.

o Cases are built into the verb itself.

Karaka

• Sanskrit uses a concept called Karaka.

• Karaka is used by Panini to construct unambiguous and grammatically correct

sentences.

• A participant involved in the action in some manner is called karaka.

• Karaka help to link the words in a sentence to the action is involved in some manner
is called Karaka

• Everything links around the action that is basic formation!

Example: Sentence Breakdown Using Karakas

• The picture in the slide shows “yantra karaka apakaroti vahana karyalaya pratahkala”

o karaka =doer .

o What is removed: "yantram" (machine/object), 2nd case *What are the

actions:""removes away"".

o Which aids (vahana/ with the vehicle) 3 rd. *"" Where (karyalaya -the place""
5th case .

o When ""Pratha h Kala -the time "" in the morning . 7th case. *So you can
jumble the words and still the meaning stands.""

Cases in Sanskrit

• Analyze 6 karakas.

• 1st Karaka - Doer (kartā):

o The cause of the action.

o Where the action is resident.

• 2nd Karaka - Object (karma):

o The locus of the result of the action.

o yantram (machine) 2nd case

• 3rd Karaka - Instrument (karana):

o Aids in the attainment of the action.

 o Makes the action easier.

o Aids instrument of work and or a vehicle for a help to get things done.

• 4th Karaka - Receiver (sampradāna):

o Where the object really goes or gets associated.

o Missing in example.

• 5th Karaka - Separation (apadāna):

o You separate it from somewhere .

o Office (Place from where you took the instrument.)

• 7th Karaka - Time and Context (adhikarana):

o The substratum on which everything is happening.

o The context in which it is working.

o ""Pratha h Kala """
• The 6th case is not a karaka and is related to owner ship ""somebody"" and that is you
can inter- connect
Sanskrit

• Cases are attached to the verb itself.

• Thus, you can put them in what ever way and all things link tightly together such that
there is no conclusion

NLP Benefit

• The structure could be read for machines and so, the words does not matter

• This can contribute for NLP systems because of very tight links that Panini offers.

3.10 Verbs and prefixes

Introduction
• This section emphasizes the importance of verbs.

• Everything revolves around the verb because action is the basis for language.

Role of Verbs

• If there is no action, there is no need for language.

• The previous discussion established the verb's importance in karakas.

• Language is only required because we engage in action.

 • Several aspects of linguistics revolve around action.

Derived Noun Roots

• Several noun roots are derived from verb roots in Sanskrit.

• This makes verbs even more important.

• Many nouns originate from verbs.

Synonyms

• Sanskrit has several synonyms for a word.

• Synonyms are derived from a dhatu (verbal root).

• Knowing the dhatus enables us to understand synonyms and their unique meanings.

Increase in power of language

• Understanding this concept and the signifcance of action significantly increases the
power of the language.

• We can choose the most appropriate synonym for the context.

Knowing Verbs
• We need to know the verbs and their origins.

Example: Fire

• Derive fire from the verb to show several possibillities.

• Several synonyms for fire possible from "kr"" root.

• Several verbs can be derived.

• vahni:

o Derived from the dhatu vah (to carry).

o Use when fire is a carrier (e.g., ahuti being offered).

• pavakah:

o Derived from the dhatu pun (to purify).

o Remove “pu", meaning to purify

o Use when discussing fire as a purifying agent.

• susma:

o Dhatu is sush (to dry or shrink).

o Use when talking about heat evaporating water or shrinking organic objects.
 • dahanah:

o Dhatu is dah (to convert or burn to ashes).

o Use when communicating fire's power to burn objects.

Appropriateness of Verbs
• Knowing these verbs is very important.

• Using the right verb conveys the meaning precisely.

• Using the correct and the appropriate Verb helps the other understand what the
message should be!

Verbs and Prefixes

• Diagram shows "kr" dhatu and its forms.

• apa-karoti, upa-karoti, ut-karoti, pra-karoti, prati-karoti, pratyupa-karoti, anu-karoti,

sams-karoti, vyakaroti, nira-karoti, adhi-karoti.

• "kr" means "to do".

Effect of Prefixes

• Attaching a prefix changes the meaning.

• Prefix and Suffix

o We have already been talking about suffix and the operation. *We are gonna
talk about Prefix now"" *Prefix and Meanings

• apa-karoti:
o apa means away.

o apa-karoti means take it away.

• upa-karoti:

o upa means closer.

o upa-karoti means bring it near (help).

• nira-karoti:

o nira means opposite.

o nira-karoti means disagree.

Power of Prefixes

• Attaching prefixes increases the power and scope of verbs.

• There are 22 prefixes in Sanskrit.

 • These can make the verb even more expressive and powerful.

Effects on Meaning

• Strengthening or Emphasizing:

o "smarati" (remembers).
o "sam-smarati" (remembers very well).

• Expanding or Improvising:

o "karoti" (does) with different prefixes (upa, apa, etc.).

• Bringing Opposite Meaning:

o "karoti" (does).

o "prati-karoti" (nullifying it).

o Gacchati (going), Aagacchati: comes (opposite direction) know that kind of

things

Prefix Usage

• It gives opposite meaning.

• You can attach more than one prefix to get the Power to be stronger.

Verbs and Language Power

• Verbs can be converted into nouns.

• Create synonyms to give expressive power.

• Attaching prefixes enhances the scope of verbs.

• Meaning, improvise, strengthen, emphasize and makes Language to be very powerful

Summary

• These features are unique to Sanskrit.

• The way we use verbs are very powerful in the Sanskrit Language.
3.11 Role of Sanskrit in natural language processing

Introduction

• Examine the role of the Sanskrit language (Panini's) in Natural Language Processing
(NLP).

Salient features of the language

• Rule-based

• Unambiguous
• Structures are modular

• You can disassemble and understand all these are interesting points""

The potential of the language

• Above mentioned points is a building block for NLP systems

Natural Language Processing (NLP)

• Concerned with processing natural language data using computers and programming
techniques.

Tasks

• Machine should be able to process heard words and repeat what it sees.
• A computer should understand human language.

• To build the bridge between language and machine

Disciplines Invovled

• Linguistics

• Computer Science

• Artificial Intelligence

Subset of Ideas in NLP

*Natural Language Generation *which will combine all ingredients ""into a machine"","" the
machine to a story.""

What is that one needs in Machine and Computers""

-generating a"""" text from a given meaning representational Logic"" Implemented through
computer program and that's called Natural language generation (NLG)""

What's Natural Language Understanding in machine?

*It should"""" distill -extract the meaning out of "" a natural language, a data""
The example,

"" I feed a sentence to the computer and"" It should give"case"" (Third case) ,""plural"",
""Singular"" and what kind of "" Verb"""

Advantage of NLC with rule based constructs

*This ""NLG -natural language generation is slightly -easier now because now if you you
give some -rules, then it will follow, particularly with all the Lexicons and Syntax""!

Inherent Advantages of Sanskrit

• Certain inherent advantages in the structure of Sanskrit and the Paninian grammar
mechanism.

• Attractive candidate for NLP and Artificial Intelligence-related work.

Question Bank for SPPU IKS Exam – Unit 3: Linguistics

Introduction to Linguistics and Language
1. What is the significance of language in communication, scientific discovery, and
societal progress?

2. Why is systematic study of language essential in science and technology?

3. Explain the relevance of Ancient Indian contributions to linguistics and language
structure.
Components and Structure of Language

4. Describe the four fundamental components of language in terms of skills and

medium.
5. What structural requirements does a language need for effective listening, speaking,
reading, and writing?
6. Define linguistics and explain its role in understanding speech, sounds, grammatical
structures, and meaning.

Panini and the Ashtadhyayi

7. Who was Panini and what is the significance of his work "Ashtadhyayi"?

8. Outline the structure and organization of Ashtadhyayi by Panini.

9. What are “sutras” in Panini’s grammar and what advantages do they offer?

10. Explain how Panini’s rules handle vocabulary, word derivation, and exceptions.
 11. Discuss the contributions of Katyayana’s Varttika and Patanjali’s Mahabhashya as
commentaries on Panini's work.

Phonetics in Indian Knowledge Systems

12. What is the role of oral tradition in the preservation of the Vedas and pronunciation?

13. Define phonetics in the Indian knowledge tradition and explain its importance in
linguistic studies.

14. What are Pratishakhyas and what issues do they address?

15. Discuss Panini’s approach to phonetics, sound origins, and articulation.

16. Explain the three types of vowel variations and the concept of ‘svaras’ in Sanskrit
phonetics.

Word Generation and Grammar

17. Explain Panini’s logic and algorithmic approach to word generation in Sanskrit.
18. Describe the steps involved in deriving a word from a verbal or nominal root.

19. How are suffixes used in forming different word cases, genders, persons, and numbers
in Sanskrit?

20. Illustrate word pattern formation using the verb roots "kru," "pat," and "gam."

21. How does recursive logic operate in compound word formation (Samasa) in Sanskrit
grammar?

Computational Elements and Algorithmic Grammar

22. How does Panini’s grammar model resemble computer language and algorithmic
structures?

23. Explain the process and logic flow in rule-based word formation in Sanskrit using
Panini’s rules.

24. What are data structures (like mnemonics) in Panini’s grammatical system and how
do they work?

Mnemonics and the Maheshwara-Sutras

25. What are Maheshwara-sutras and how are they used in Sanskrit grammar?

26. How do mnemonic devices streamline rule application and data representation in
Panini’s grammar?

27. Give examples of how Maheshwara-sutras are used to derive sets and rules in
Sanskrit.

Recursive Operations and the Kaprekar Constant

 28. Describe the concept of recursion in Panini’s Ashtadhyayi and its relevance to
computational linguistics.

29. What is Samasa and how is recursive logic applied in compound word creation?

30. Explain the introduction and use of the Kaprekar constant (6174) as a recursive
operation in this context.

Sentence Formation and Karakas

31. What is the basic logic behind sentence structure in Sanskrit as per Panini?

32. Explain the role of the verb in the construction of sentences and the concept of
‘karaka’.

33. How does word order flexibility in Sanskrit benefit communication and Natural
Language Processing (NLP)?

34. How are cases and karakas attached and interpreted in Sanskrit sentences?

Verbs, Prefixes, and Language Power

35. Why are verbs central to word formation and meaning in Sanskrit?
36. Describe the impact of prefixes on verbs and meaning creation.

37. How are several noun roots and synonyms derived from verbs in Sanskrit?

38. Explain with examples how the use of prefixes can change and strengthen the
meaning of verbs.

Sanskrit and Natural Language Processing (NLP)

39. What features of Sanskrit and Panini’s grammar make it highly suitable for NLP and
AI applications?

40. Define Natural Language Generation (NLG) and Natural Language Understanding
(NLU) in the context of Sanskrit.

41. How does the modular, rule-based, and unambiguous structure of Sanskrit grammar
help computational tasks?