The Problem In modern philosophy, knowledge has often been treated as the domain of truth, evidence, reasoning, and objective inquiry, while power has been treated as the domain of the state, repression, command, institutions, or social control. Michel Foucault questions this clean separation—especially in forms of knowledge that create categories concerning human beings, crime, illness, sexuality, education, normality, madness, and population. Foucault's question is not that 'truth is impossible' or that 'knowledge is false'. His question is: which institution makes what into an object of knowledge? Who receives the authority to speak? Through what measurements, examinations, records, classifications, and normalizations does a person become an object of knowledge? And how does this knowledge in turn become an instrument for controlling conduct? A school examination produces knowledge about students—marks, rank, attendance records, ability, and classifications such as 'average' and 'weak'. This knowledge also has administrative consequences—pass/fail, admission, scholarship, discipline. Here knowledge and power no longer remain wholly separate domains. A hospital creates charts, diagnoses, test results, and risk categories concerning patients; a prison creates prisoner files, conduct records, and risk assessments; a state creates a census, identity records, and demographic categories. Foucault's question of power/knowledge centres on this reciprocity. But one dangerous over-conclusion can be drawn from this critique—'knowledge is connected with power, therefore all knowledge is arbitrary.' Parallel Philosophy rejects this conclusion. Power can influence the production, distribution, visibility, funding, institutional location, and use of knowledge; but reality, evidence, replication, and independent inquiry provide distinct controls on knowledge. The problem of this chapter is therefore: how closely are knowledge and power connected? How does power produce or structure knowledge? How does knowledge enable power? And while criticizing this relation, how can objective truth, scientific reliability, and moral responsibility be preserved? Central Thesis The central thesis of this chapter is that knowledge and power have a mutually constitutive relation in many human institutions, especially when knowledge measures, classifies, normalizes, and administers individuals or communities. Foucault's concept of 'power/knowledge' says that power is not merely an external distortion of knowledge; at times relations of power generate an entire field of questions, categories, records, expertise, and examination. The reverse side of this relation is also important—knowledge makes power more fine-grained. Through observation, data, examination, diagnosis, classification, files, and norms, institutions see individuals' conduct, compare it, and intervene in it. Parallel Philosophy accepts this analysis but adds one explicit limit: 'Power influences knowledge; it does not make truth arbitrary.' Table 65 — Multi-evidential rationalism proposes a power-aware epistemology—examine the institutional location of the source, and also examine the independent evidence for the claim. Instrument of knowledge Power relation Category produced Parallel criterion Examination Measurement / ranking Qualified / unqualified Standard + appeal Medical test Expert intervention Normal / illness Scientific evidence Archive / record Documentary visibility Case / file Source criticism Census Administrative counting Population category Definition / rights Surveillance Visibility of conduct Risk / deviation Proportionality / privacy Statistics Average / distribution Norm / deviation Description ≠ moral value Algorithm Automated classification Score / rank Audit / appeal Table 36.1 — Knowledge production, institutional power, and Parallel examination. Principal Arguments First argument—Power is not only prohibitive. In Foucault's analysis, power is also productive: it can produce frameworks of conduct, expertise, classifications, normalities, skills, disciplines, and fields of knowledge. Second argument—Knowledge is not merely an instrument of power. Power can be blind without knowledge; modern institutions develop more subtle forms of control through observation, record-keeping, measurement, and expert judgment. Third argument—The examination is an exemplary case of power/knowledge. An examination produces truth-claims about a person and at the same time changes the direction of that person's conduct. The student studies because there is an examination; the institution decides because results are available. Fourth argument—Normalization is a kind of power different from law. Law may say 'permitted/prohibited'; normalization creates a continuum such as 'average/deviant', 'competent/weak', or 'healthy/abnormal'. Fifth argument—Records give a person a 'documentary reality'. School records, medical files, prison dossiers, credit histories, and digital profiles can become enduring bases for decisions about a person. Sixth argument—Foucault's power is not merely a central conspiracy. Diverse institutions, rules, professions, norms, and local practices can together produce an arrangement of power with no single master planner. Seventh argument—Relations of power also open the possibility of resistance. A category, diagnosis, identity record, curriculum, or policing method can be contested; affected persons and communities can challenge definitions. Eighth argument—It is necessary to examine the institutional origin of knowledge, but it is equally necessary to avoid the genetic fallacy. A claim is not automatically false because it came from a bad institution; it is not automatically true because it came from a prestigious institution. Ninth argument—In scientific knowledge, relations of power can influence funding, publication, research agendas, access, patents, professional hierarchies, and data ownership. But experiment, replication, measurement, peer criticism, and independent confirmation provide distinct criteria of reliability. Tenth argument—In the human sciences, classification can be reflexive or self-affecting. Labels such as 'criminal', 'gifted', 'disabled', or 'at risk' can alter the behaviour of persons and institutions; the category itself can produce social consequences. Eleventh argument—A statistical norm is not a moral norm. To be 'average' is not to be 'superior'; to be 'deviant' is not to be a 'defective human being'. Dignity-based ethics keeps descriptive classification and moral worth distinct. Twelfth argument—Power/knowledge analysis raises the question of the place of truth: 'Who is able to speak truth?' 'Which institution is recognized as a legitimate knower?' But the Parallel answer adds: 'How well does the claim withstand independent evidence?' Thirteenth argument—Power influences source survival in historiography. Rulers, courts, land officials, universities, the press, and archives may leave more documents; oral and marginal experiences may leave fewer. This is archival asymmetry, not the impossibility of history. गजेन्द्र ठाकु रक समानान्तर दर्शन — खण्ड २ Fourteenth argument—The ethical evaluation of a knowledge–power relation depends on its effects. Disease classification can provide medical help; the same classification can also create stigma. The question is not only 'Is there classification?' but 'For what purpose, by whose authority, and to whose benefit or harm?' Fifteenth argument—Parallel Philosophy proposes power-aware objectivity: acknowledge the power relations of the source, make procedures transparent, include counterpositions, maintain independent scrutiny, and make conclusions revisable. Pūrvapakṣa The first form of the pūrvapakṣa says that the concept of power/knowledge is itself excessively broad. If power exists in every social relation, then 'power' may say nothing analytically specific. The second pūrvapakṣa is realist: while narrating the social history of scientific truth, we may confuse the social causes of knowledge with the cause of truth itself. The earth revolves around the sun—this is not a power-produced fact. The third pūrvapakṣa is liberal: Foucault criticizes institutions, discipline, and normalization so strongly that the legitimate functions of law, education, public health, and administration can recede from view. The fourth pūrvapakṣa is ethical: if power is everywhere, how clearly can oppressive power be distinguished from legitimate authority? The fifth pūrvapakṣa concerns Indian comparison: to call rājadharma, daṇḍanīti, scriptural authority, guru–disciple traditions, or caste-based distribution of knowledge directly 'Foucauldian power/knowledge' would be historically careless. Uttarapakṣa Reply to the first pūrvapakṣa—Saying 'power is everywhere' is not enough; identify the specific mechanism: who observes, who makes the record, who determines the rule, who applies sanction or approval, who hears an appeal? Specificity should be a criterion of power analysis. Reply to the second pūrvapakṣa—Power can be a cause in the social production of knowledge, not the ontological cause of the fact. Scientific claims should be tested against evidence from the world. Reply to the third pūrvapakṣa—Critique of discipline is not opposition to institutions. Schools, hospitals, censuses, and law may be necessary; the question concerns proportionality, rights, transparency, consent, and mechanisms of correction. Reply to the fourth pūrvapakṣa—Legitimate authority requires public reasons, protection of rights, proportionality, accountability, appeal, and democratic control. Mere influence does not equal legitimate power. Reply to the fifth pūrvapakṣa—Comparison with Indian traditions should remain at the level of problem-dialogue: the social distribution of knowledge, the authority of pramāṇa, access to śāstrārtha, educational institutions, and the social consequences of classification. It should not claim historical identity. Indian Dialogue Nyāya analyzes the validity of pramāṇas independently of power—the knowledge-producing conditions of perception, inference, comparison, and testimony. In this respect Nyāya offers a strong counterposition to Foucauldian power/knowledge analysis: a truth-claim is not valid merely because of social authority. In śabda-pramāṇa, the reliability of the speaker, sentence meaning, and the knowable object show that 'who speaks?' matters, but 'who has power?' is not the only criterion. Mīmāṃsā raises complex questions of scriptural authority, sentence meaning, and injunction; it is not directly identical with Foucauldian institutional genealogy. The point of dialogue is: which text or speaker receives epistemic privilege, and what is the rationale for that authority? The Buddhist pramāṇa tradition examines sources of knowledge and the limits of perception/conceptualization and inference. Power analysis is not a substitute for these epistemological questions. Kauṭilya's Arthaśāstra can provide a historical Indian example of the relations among state, counting, punishment, espionage, administration, and knowledge of populations and resources; but this does not establish sameness or identity with Foucault's modern disciplinary power. Dharmashāstric classifications possessed normative authority in society; their historical effects should be examined according to region, period, and practice. A scriptural prescription is not automatically identical with actual social conduct. Śāstrārtha was an institution for public or semi-public testing of knowledge, but participation was not universally equal. The Parallel perspective examines logical achievement and social access separately. Conclusion of the Indian dialogue—Both the validity of knowledge and the social power of knowledge are necessary questions; neither substitutes for the other. Mithila's Parallel Perspective In Mithila's intellectual history, scholarly traditions, pañjī, educational institutions, land records, courtly patronage, religious prestige, print, and modern academic institutions have influenced the visibility of knowledge. Documents such as pañjī or genealogies provide knowledge of social classification; at the same time they may themselves be part of the practice of classification. Therefore they should be read not merely as neutral records, but as social documents. Printed books, journals, curricula, and awards determine which authors, languages, and genres receive canonical visibility. This is a knowledge–power question. But a less-published author is not therefore automatically greater or truer. Standardization of Maithili is a useful power–knowledge process for education and publication. The questions are: who makes the standard? What place do regional forms have? Is the revision process open? Parallel historiography states archival inequality explicitly. 'More sources' does not mean 'more historical importance'; 'fewer sources' does not permit fabricated compensation either. Multi-evidential rationalism can combine institutional sources with oral history, folk songs, private letters, regional publications, archaeology, digital archives, and statistical evidence. Dignity-based ethics examines the human consequences of classification. Classifications based on caste, gender, language, occupation, or education may be administrative facts; they are not measures of moral human worth. Table 66 — Polycentric democracy offers an answer to the distribution of knowledge–power—no single academy, state, university, editor, platform, or algorithm is the final owner of a field of knowledge; mutual criticism and appeal are necessary. Question Institution Record Classification Power Evidence Appeal What is to be Does the claim Is revision Who measures? What is recorded? Which category? Who decides? known? hold? possible? Figure 36.2 — Examination of knowledge–power: from question through institution, record, classification, decision, evidence, and appeal. Contemporary Applications In artificial-intelligence (AI) knowledge systems, the power/knowledge question is direct. Who selects the training data? Which sources are raised by ranking? Who determines safety labels? How does user feedback alter model behaviour? Alongside all of these, factual accuracy should be examined separately. In algorithmic hiring, credit, and policing, a score is both a knowledge-claim and a basis for a power decision. A Parallel audit examines data quality, feature choice, historical bias, false positives, explanation, appeal, and disparate impact. In medicine, a diagnostic category can provide life-saving knowledge; it can also affect insurance, stigma, employment, or surveillance. Medical validity and social consequence should be examined separately. गजेन्द्र ठाकु र In education, standardized tests make comparison possible; they can also create curriculum narrowing, coaching advantages, and labeling effects. A score is not the whole of a person's ability. Digital platforms create categories for content moderation—spam, harassment, misinformation, adult content. Definitions are necessary, but transparent rules, context sensitivity, and appeal are also necessary. For the state, census and identity data are necessary for policy and welfare; they also carry risks of privacy invasion, exclusion, misclassification, and surveillance. Data minimization and purpose limitation should be elements of power-aware design. In science policy, research funding can influence the research agenda. A less-funded question is not therefore false; a more- funded question is not therefore true. Evidence quality should be examined independently. In history and journalism, archive/search ranking can over-represent what is easy to find. The researcher's duty is to map both visible evidence and missing evidence. Chapter Conclusion Knowledge and power are not wholly separate domains, especially when institutions measure, classify, record, and normalize human beings. Foucault's analysis of power/knowledge shows that examination, surveillance, documentation, and norms both produce knowledge and control conduct. But 'power influences knowledge' does not entail that 'all knowledge is nothing but power'. Scientific, historical, and everyday claims can be tested through independent evidence, replication, source comparison, and the resistance of reality. Parallel Philosophy proposes power-aware objectivity—examine the institution of the source, examine the consequences of classification, open the counterposition, retain independent evidence, and make appeal and revision possible. In Indian philosophy, debates about pramāṇa provide a necessary counterposition for critique of power/knowledge: authority does not equal truth; at the same time, the social distribution of epistemic authority is not exempt from criticism. The final maxim of the chapter is: 'Power influences knowledge; it does not make truth arbitrary. Examine knowledge—and also examine the institutions that produce knowledge.'