Metaphor and Metonymy: Semantic Concepts Explained

This section covers conceptual blending theory application and cross-linguistic analysis. Conceptual blending theory (Fauconnier and Turner) explains how concepts are formed through integration of multiple input spaces into a blended space that creates new meaning. For the 'successful woman' concept, this approach allows analysis of how the concept combines elements from different domains. The theory is considered ideal for studying the ontological nature of unified conceptual content. Cross-linguistic analysis reveals semantic differences: in Ukrainian, 'невдача' (failure) is associated with misfortune; in English, 'loser' refers to someone who lost a competition; in French, 'malchance' refers to misfortune through causes independent of the person. These differences reflect how the nature of 'success' is semantically different across languages, demonstrating the importance of cross-linguistic analysis in conceptual research.

Conceptual blending explains how humans create novel conceptual combinations through input spaces, vital relations, and selective integration into blended spaces with emergent structure. The theory accounts for diverse phenomena including morphological blends ('snack accident'), compounds ('land yacht'), predication constructions ('drive at a safe speed'), and argument structure innovations ('she Gandalf me'). A detailed analysis of 'If I were my brother-in-law, I'd be miserable' demonstrates how blending combines distinct identities to produce genuinely new meanings not present in either input space.

Conceptual integration (or conceptual blending) is a cognitive process that combines two or more input spaces into a new blended space through compression of vital relations (such as time, identity, space, role, and cause-effect), creating emergent structures that weren't present in either input; unlike conceptual metaphor which involves a single source-to-target mapping, conceptual integration can involve multiple input spaces and generates novel meanings that arise from the interaction of these inputs, operating pervasively across grammatical domains including negation, modality, causation, and attributive constructions.

Conceptual blending (Fauconnier & Turner) applies to combining construction meanings, both conventional and creative. Input spaces map to blended spaces through cross-space projection, explaining how 'feel guilty' and 'have a conscience' combine to 'have a guilty conscience.' This framework accounts for creativity within constraints. The constructicon integrates all constructions, with some having strong lexical bias (like 'give' dominating ditransitive) and others minimal bias. This integration enables pedagogical applications, helping teachers show how each construction contributes to sentence meaning and form, bridging theoretical analysis with practical language instruction.

Conceptual Blending Theory, developed by Mark Turner and Gilles Fauconnier over 30 years, proposes that advanced human cognition relies on a fundamental cognitive process called 'blending'—the integration of multiple mental spaces into emergent structures that enable creative thinking, language comprehension, and problem-solving across diverse domains including mathematics, art, science, and AI. Unlike traditional cognitive approaches that treat different mental operations as separate, blending theory argues that humans use the same core cognitive mechanism for all advanced cognitive tasks, from understanding metaphors to planning futures, making it a foundational framework for a unified science of human cognition.

Metaphor and metonymy are both figures of speech based on substitution but differ fundamentally in mechanism. Metaphor substitutes one thing for another based on similarity or resemblance, allowing transformation of meaning. The example 'the sky is full of cotton' substitutes clouds for cotton due to shared characteristics like texture and color. Metonymy substitutes based on continuity through cause-and-effect relationships, spatial-temporal connections, or creator-work relationships. Metonymy predominates in advertising because it appeals to common understanding of cause-and-effect relationships. Examples include Trident gum (substituting gum for toothbrush bristles), Adidas football boots (representing players through their boots), and Bimbo bread (showing milk to represent bread made with milk). Metonymy is more prevalent in visual media because it requires less cultural knowledge to understand.

This segment teaches students how to identify and analyze rhetorical devices in political writing. The instructor explains 'استعارة مكنية' (metaphorical chest) where the nation is described as having a chest that experiences distress, and 'تلميح' (metaphorical suggestion) where the nation's inability to find its path is implied rather than stated directly. The segment also covers 'استعارة مكنية' in describing leadership failure, where the nation is compared to a caravan led astray by deceitful guides. These devices allow authors to convey complex political ideas vividly while maintaining rhetorical elegance.

Metaphor (istighara) is a type of simile where one component is omitted. There are two types: 1) Istighara Makhniyah - omits the mushab bihi (the thing being compared to), e.g., 'The soldier roars' (comparing the soldier to a lion by mentioning only the lion's characteristic 'roar'); 2) Istighara Tasrihiyah - omits the mushab (the thing being compared), e.g., 'We crossed the black channel' (comparing soldiers to black ants by mentioning only 'black'). Metonymy (kinayah) is a word used to mean something other than its literal meaning, with three types: 1) Kinayah 'an Sifah (about a characteristic) - describes a characteristic indirectly, e.g., 'The fire before us did not go out' (describing generosity); 2) Kinayah 'an Mawsuf (about the described object) - refers to the object itself indirectly, e.g., 'The owner of the whale' (referring to Prophet Yunus); 3) Kinayah 'an Nasbah (about relationship) - describes something through an indirect relationship, e.g., 'The glory is in your clothes' (describing glory through clothing).

Advertising uses metaphors (relating one thing to another to communicate positive values indirectly) and metonymies (accessing complex domains through partial information) to create effective campaigns, but these linguistic devices can be interpreted differently across cultures due to varying experiences, making linguistic research essential for anticipating consumer inferences and avoiding cross-cultural misunderstandings in global marketing.

Metonymy is a rhetorical device where a shorthand term stands in for a literal term. Apple's '1,000 songs in your pocket' uses 'pocket' instead of 'media player' because it's more visual and punchy. While technically incorrect (songs aren't literally in pockets), the metaphor creates a more memorable and emotionally resonant message.

Computational linguistics requires combining traditional linguistics with computer science and artificial intelligence. A typical educational path includes studying linguistics (such as Germanic philology), followed by specialized training in computational linguistics and artificial intelligence. Key research areas include lexical semantics, lexicography, syntax, and knowledge representation. Natural Language Processing (NLP) is an interdisciplinary field combining computer science, AI, and linguistics, with applications including chatbots, voice recognition, opinion mining, and machine translation. Machine learning in NLP works through pattern recognition rather than true understanding - text is converted to numerical representations, and algorithms search for patterns enabling generalization from training data. Linguistic resources face integration challenges: isolation, format heterogeneity, and structural inconsistency make it difficult for machines to automatically combine and navigate data.

Natural Language Processing (NLP) is a field at the intersection of computer science, AI, and computational linguistics, concerned with computer-human language interactions. Unlike artificial languages with predefined grammars, natural languages evolved automatically. NLP faces immense challenges due to linguistic diversity—over 6,000 official languages worldwide, with India alone having ~19,000 languages including dialects. The Turing Test evaluates machine intelligence through text-based communication. Fundamental challenges include: (1) Semantic vs literal meaning differences, as in headlines where 'death' might mean 'trump card'; (2) Syntactic ambiguity where identical structures produce different interpretations; (3) Punctuation's critical role in meaning, as shown by 'Let's eat Grandma' versus 'Let's eat, Grandma'; (4) The famous 'Buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo' example demonstrating extreme morphological richness where each 'buffalo' has different meanings (animal, state, verb).

Computational linguistics, or natural language processing (NLP), involves programming computers to understand and generate human language through a multi-step process: converting input (speech, text, or images) into digital format, analyzing word meanings and relationships, interpreting meaning, performing useful tasks like translation or question-answering, and generating human-readable output. Despite advances using neural networks and machine learning, computers still struggle with language because they lack true understanding—they follow patterns rather than comprehension. Additionally, NLP systems face significant biases from training data, including historical bias (reflecting real-world prejudices), representation bias (underrepresentation of minority languages and signed languages), measurement bias (using inappropriate training materials), aggregation bias (failing across language varieties), and deployment bias (misuse of tools). Addressing these challenges requires ongoing research and ethical consideration in developing language technologies.

Natural language processing represents one of the most difficult areas for AI to master. The ambiguities of syntax and context restrict present systems to very limited applications. A simple geographic question depending on phrasing can lead to multiple interpretations, requiring exhaustive deconstruction of sentences before the program determines the most likely meaning, and may even reject unusual phrasings of the same query.

Computational linguistics, also called natural language processing (NLP) or human language technology, is best understood as a subfield of artificial intelligence rather than linguistics. Language can be studied from eight perspectives: linguistics (systematic conventions), cognitive science (mental processing), computational linguistics (treating language as data), social linguistics (communication between people), and anthropological linguistics (worldview categorization). NLP applications include spell checking, speech recognition, sentiment analysis, document classification, hate speech detection, language documentation, speech synthesis, text summarization, question answering, intelligent assistance, and machine translation. Before the 21st century, NLP relied on brittle rule-based systems requiring expert-crafted rules. Around 20 years ago, machine learning transformed NLP by using massive data rather than hand-crafted rules, with modern systems using neural networks with billions of parameters that remain opaque even to developers.

Cognitive poetics focuses on the relationship between text structure and its effects on readers. Literary texts impact audiences primarily through emotional effects that make readers think, feel sad, or happy. This occurs because readers enter the world of the characters and become immersed in the literary narrative. The theory encompasses key concepts: representation of the real world in literature, the nature of literary creativity, and the relationship between everyday language and literary language. Creativity involves deviation from cognitive frameworks and presuppositions, transforming ordinary language into artistic expression. The foundational works include Lakoff and Johnson's 'Metaphors We Live By' (1980), Turner's 'The Death of the Mother,' and Tsur's 'The Poetics of Cognitive Linguistics.'

Cognitive Stylistics or Cognitive Poetics is a recent development in Stylistics that applies cognitive linguistics, cognitive psychology, and artificial intelligence to literary analysis. At its most basic level, it is 'all about reading literature'—the psychological process involved in reading. It highlights aspects of creating that literature consumers operate when processing literary texts, including building knowledge, forming concepts, developing schemas, and improving memory. The approach combines cognition (mental processes) with Poetics (the craft of literature) to understand how readers mentally process and construct meaning from texts.

Cognitive Poetics emerged in the late 1980s-early 1990s as scholars from linguistics and literary studies began applying cognitive science methods to analyze literature, initially working in isolation before discovering a global community through conferences and internet connectivity; this field combines knowledge of language structure with insights into how readers mentally process literary texts, using tools like metaphor analysis and conceptual blending to understand literary meaning.

Cognitive poetics is the application of cognitive science principles to the study of literature, focusing on how readers construct mental representations (text worlds) when engaging with texts, and how stylistic features guide readers in activating relevant background knowledge to interpret narratives; unlike structuralist approaches, it emphasizes the dynamic, reader-centered process of meaning-making rather than fixed structural elements.

Reading is not passive but an active process where readers decode messages, distinguish truth from fiction, identify irony, and organize narrative space. Cognitive poetics is an interdisciplinary approach combining psychology, linguistics, and cognitive science to study how readers process texts. This methodology has practical applications for creating and analyzing artistic texts, moving beyond the misconception that all stories are already told and only need to be copied.