The Problem One of the oldest and most difficult questions in the philosophy of language is: what, after all, is the meaning of a word? Is the meaning of the word ‘tree’ the tree before us? A general concept of a tree? The social use of the word ‘tree’? Or is the meaning of ‘tree’ the network of relations in which words such as ‘plant’, ‘forest’, ‘fruit’, ‘wood’, ‘living’, and ‘root’ are connected? If meaning is only an object, how can terms such as ‘unicorn’, ‘justice’, ‘democracy’, ‘pain’, ‘emptiness’, and ‘number’ have meaning? If meaning is only use, what becomes of incorrect use, false statements, and the relation to the external world? If meaning is only relation, how does the network of relations ultimately connect with the world, the speaker, and social practice? Frege's distinction between sense and reference shows that two expressions pointing to the same object can have different cognitive value. Russell uncovers the logical form of descriptions. Later Wittgenstein directs us to look for meaning in use. Modern externalist approaches show that meaning is not confined to the speaker's mind; the linguistic community and the actual environment also play a role. Indian theories of linguistic meaning raise the question in different ways: Nyāya through the word–referent relation, universal– individual, secondary signification, and verbal testimony; Mīmāṃsā through sentence meaning, speaker's purport, and word- power; the grammatical tradition through the unity of the sentence; and Buddhist apoha through the negative delimitation of meaning. The problem of this chapter, therefore, is not the simple question, ‘Which theory is correct?’ The aim is to distinguish the different dimensions of meaning: reference, conceptual presentation, use, social rule, relational difference, sentence context, and actual environment. Then we can ask which dimension is necessary for which kind of claim. Central Thesis The central thesis of this chapter is that it is difficult to reduce the meaning of a word to a single element. Meaning is multilayered: sometimes objective reference is essential, sometimes cognitive presentation, sometimes social use, sometimes linguistic relation, sometimes sentence context, and sometimes community and environment. Parallel Philosophy does not turn this plurality into arbitrary semantic relativism. The meaning of a word does not depend entirely on private choice; public linguistic practice, historical use, objective reference, definition, textual context, and evidence together establish its limits. The formula, therefore, is: ‘Meaning is not merely an object, not merely use, and not merely relation. Meaning is a multilayered structure of linguistic form, social use, and relation to reality.’ In this multilayered model, the meanings of different words may be constituted in different proportions. In a name such as ‘Ganga’, reference is primary; in a term such as ‘game’, use and family resemblance; in ‘justice’, conceptual and institutional contestation; in ‘pain’, both private experience and public language; and in ‘electron’, scientific theory, instrumentation, and reference to reality. Table 31 — Five Principal Dimensions of Meaning Dimension Principal question Example Limitation Reference alone does not explain Reference What does the word point to? “Ganga” cognitive difference Does not fully encompass use and Mode of presentation How is the object presented? “the morning star” institution Truth/reality remains a separate Use What does the word do? “game”, “promise” question How is it distinguished Relation “red”, “colour” Relation to the world is necessary from/connected with other terms? Where, by whom, and how is it A fixed essence may be less Context/community “bank”, “pain” used? prominent Table 19.1 — Five dimensions of meaning; no single dimension is sufficient for every word. Principal Arguments First argument—reference is an important part of meaning. The word ‘Patna’ points to an actual place; if the reference is wrong, the factual value of the statement can change. But reference alone is not sufficient: the same object may be presented differently by different expressions. Second argument—Frege's sense–reference distinction shows that ‘the morning star’ and ‘the evening star’, although referring to the same planet, do not have the same cognitive significance. Thus a mode of presentation is necessary in meaning. Third argument—Russell's theory of descriptions shows that an expression that looks like a name may in fact contain complex claims of existence, uniqueness, and predication. Therefore, a superficial word–object relation is not sufficient. Fourth argument—Later Wittgenstein's account of use brings words back into life. The meaning of ‘game’ may be understood not through a single shared essence but through overlapping varieties of use. Fifth argument—if meaning is treated as nothing but use, factual reference may be weakened. A community may meaningfully use the sentence ‘the sun revolves around the earth’ for a long time, yet the statement can be factually false. Meaning and truth are distinct. Sixth argument—the relational view shows that a word's meaning is shaped through differences from and relations to other words. ‘Hot’ forms a semantic field in relation to ‘cold’, ‘mother’ to ‘child’, and ‘buying’ to ‘selling’. Seventh argument—but a network of relations alone is not sufficient. If an entire network of words is cut off from the world, an imaginary language may remain internally coherent while lacking external reference. Some forms of meaning must be connected with actual objects, actions, and social institutions. Eighth argument—externalist thought offers an important correction: meaning is not merely the speaker's mental state. The meaning of a natural-kind term such as ‘water’ can be shaped by the actual substance and by expert and communal knowledge. Ninth argument—the meaning of a word is not wholly separable from the sentence. ‘Bank’ may mean the bank of a river or a financial institution; sentence context selects the meaning. The Mīmāṃsā concepts of expectancy, semantic fitness, proximity, and purport shed independent light on this problem. Tenth argument—meaning is also historical. The use of terms such as ‘computer’, ‘network’, ‘caste’, ‘nation’, and ‘citizenship’ changes over time. A dictionary may provide a static picture; living meaning develops through history, institutions, and use. Eleventh argument—the plurality of meaning is especially important in translation. Merely transliterating an English technical term does not convey its meaning. A Maithili equivalent, its range of use, definition, examples, and, where necessary, the original English term together constitute philosophical translation. Twelfth argument—the model ‘one word = one meaning’ is inadequate for digital search and artificial intelligence (AI) as well. A system must combine context, co-occurrence, world knowledge, speaker intention, and document context; learning relations from a language model does not by itself verify actual reference. गजेन्द्र ठाकु रक समानान्तर दर्शन — खण्ड २ Pūrvapakṣa The first form of the pūrvapakṣa says that a multilayered model of meaning is excessively loose. If reference, use, relation, community, and context are all parts of meaning, where is the clear theory? The second pūrvapakṣa is referentialist: the principal function of language is to designate the world; the remaining elements are psychological or pragmatic descriptions. Adding more levels creates unnecessary complexity. The third pūrvapakṣa is use-theoretical: talk of objective reference cannot ultimately move outside linguistic use. We discuss ‘objects’ within language-games too; therefore use is primary. The fourth pūrvapakṣa is relationalist: the value of a word arises from differences within a system; external objects alone do not give language its structure. The fifth pūrvapakṣa comes from an Indian perspective. If Nyāya, Mīmāṃsā, Bhartṛhari, or apoha are read through the modern framework of ‘reference/use/relation’, original Indian concepts may be forcibly fitted into modern categories. Uttarapakṣa Reply to the first pūrvapakṣa—the multilayered model is not looseness but a division of analytical labour. For each claim we should ask which dimension is decisive: name-reference? sentence context? social use? classificatory relation? scientific reality? These questions make the analysis clearer. Reply to the second pūrvapakṣa—reference is necessary, but the model ‘meaning = object’ does not adequately explain non-existent terms, abstract nouns, different cognitive values with the same reference, or speech acts. Reply to the third pūrvapakṣa—use is fundamental to understanding meaning, but factual claims made on the basis of meanings constituted through use can still be tested against the world. Language-games do not abolish the resistance of reality. Reply to the fourth pūrvapakṣa—relationality is powerful, but networks of relation are used in actual practices involving bodies, objects, institutions, and environments. Complete internalism can turn meaning into a closed system. Reply to the fifth pūrvapakṣa—Indian traditions should not be forced into modern categories. First clarify each tradition's own terms, arguments, and aims; modern comparison should occur only at the level of dialogue around shared problems. Indian Dialogue Nyāya connects meaning with objects, classes, relations, and the production of knowledge through the word–referent relation, universal–individual, qualifier–qualificand, secondary signification, memory, and verbal testimony. It therefore presents a powerful realist model of the word–object relation. Mīmāṃsā highlights the importance of context, purport, expectancy, semantic fitness, and proximity in questions of word-power and sentence meaning. From this perspective, a word may be meaningful, yet complete communication is not possible without sentence-level cognition. Bhartṛhari emphasizes the unity of the sentence and the wholeness of linguistic experience. Separate word meanings may be products of later analysis; linguistic cognition may originally be more holistic. This is not the same as modern use theory or holism, but an independent philosophy of language. The Buddhist theory of apoha understands meaning through ‘exclusion of the other’. The meaning of ‘cow’ is not an eternal universal entity but is connected with a process of delimitation from the non-cow. This can enter into dialogue with relational and differential questions, but it is not a precursor of modern structuralism or deconstruction. The Indian dialogue supports the conclusion that ‘meaning’ is not a linear subject. Object, universal, sentence, context, purport, exclusion, speaker, and cognition are different centres selected by different traditions. Mithila's Parallel Perspective Mithila's Navya-Nyāya tradition makes the relational limits of claims extremely precise in questions of meaning. Analyses such as ‘which property is in which object, through what relation, and with what delimitor?’ make the word–object relation more complex than simple naming. Parallel Mithila combines this precision with a later sensitivity to use: technical definitions should be clear, but living Maithili usage should also be observed. A difference between dictionary meaning and popular usage is not automatically an error; at times it may indicate history, region, or social experience. Terms such as ‘Mithila’, ‘Maithili’, ‘folk’, ‘tradition’, ‘śāstra’, ‘modern’, ‘caste’, and ‘community’ carry enormous weight in historical writing. It is dangerous to draw sweeping conclusions without specifying their meaning in relation to period, source, and institutional use. For this reason the Parallel method proposes five questions for every important term: What is its reference? Where is it used? What is its relation to other terms? What is the historical period? What is the evidence? The goal of philosophical writing in Maithili is not merely to find equivalents for English terms, but to build philosophical usability. The word should be meaningful within the reader's linguistic life, its logical limits should be clear, and where necessary the original foreign term should remain in parentheses as an aid. From this perspective language also becomes democratic. Specialist precision is necessary, but knowledge should not remain confined within the private terminology of experts; public linguistic transformation is an intellectual responsibility of philosophy. Table 32 — Multilayered Test of Word Meaning Word Reference Use Relation Context Test To what/whom does What work does it How is it connected Meaning + fact + Which term? Period/community/sentence? it refer? perform? with other terms? evidence Figure 19.2 — A sequence for testing a question of meaning by dividing it into reference, use, relation, context, and evidence. Contemporary Applications Artificial intelligence (AI) language models learn vast networks of word co-occurrence and relation. For this reason they can produce contextually meaningful answers; however, co-occurrence relations do not automatically guarantee actual reference. Fabricated quotations or non-existent sources are examples of this limitation. In a search system, a name such as ‘Jagannath Mishra’ may refer to several persons. To determine the correct meaning, name, profession, period, place, document, and user intention have to be combined. In law, a word may have a distinct institutional meaning. ‘Person’, ‘consent’, ‘harm’, and ‘ownership’ acquire legal definitions different from everyday use; yet courts also consider social usage and purpose. In history, projecting modern words onto the past is dangerous. Today's meanings of ‘nation’, ‘caste’, and ‘language’ do not automatically apply to medieval sources. Historical determination of meaning should be based on usage in the sources. In medicine, ‘pain’ has meaning at four levels: private experience, public language, clinical criteria, and social trust. The dominance of any one level can distort either the patient's experience or medical judgment. In political discourse, words such as ‘freedom’, ‘security’, ‘tradition’, ‘religion’, ‘minority’, and ‘the people’ form networks of emotional, institutional, and legal meanings. Before argument begins, the domain of use of these words must be clarified. Chapter Conclusion It is difficult to reduce the meaning of a word in a linear way to object, use, or relation. Reference, mode of presentation, use, relation, context, community, history, and actual environment can all be distinct dimensions of meaning. गजेन्द्र ठाकु र Frege distinguishes reference from cognitive presentation; Russell uncovers the logical form beneath surface descriptions; Later Wittgenstein shows the social use of meaning; externalism adds the role of the actual environment and community. Indian Nyāya, Mīmāṃsā, Bhartṛhari, and Buddhist apoha present independent theories of meaning. It is inappropriate to turn them into precursors of modern terminology; dialogue around problems is more fruitful. The formula of Parallel Philosophy is: meaning is neither completely closed nor completely free. Public use sets limits to a word; the world and evidence constrain factual claims; history can change meaning. The chapter's final formula is: it is not enough to ask, ‘What does the word mean?’ Ask instead: ‘What does it refer to? In what context? What does it do? To which words is it related? From what history does it come? And what is the evidence for the facts asserted on the basis of this meaning?’