Human communication has long relied on natural idioms that evolved organically over millennia through communal interaction, cultural exchange, and phonetic drift. In contrast, an artificial language represents a consciously devised communicative system characterized by deliberate grammatical codification, intentionally constructed lexicons, and systematic semantic mapping. By interrogating how artificial languages are formulated, theorists and cognitive scientists gain unparalleled access into the architecture of the human mind, the formal constraints of syntax, and the operational boundaries of symbolic logic.
Artificial Language
1. Concise Definition
An artificial language—often designated within linguistics as a constructed language or conlang—is a communication system whose phonology, morphology, syntax, and vocabulary are consciously designed and engineered by an individual or a committee, rather than developing organically through cultural evolution. Unlike natural languages, which adapt over generations through spontaneous societal usage, an artificial language emerges as a discrete artifact created for philosophical inquiry, international auxiliary communication, artistic immersion, or computational processing.
In formal linguistics and cognitive science, the term encompasses both spoken human-like linguistic frameworks and formal symbolic systems used in mathematics, logic, and computer science. When applied to human speech, it delineates systems that intentionally replicate or alter the typological universals of natural languages. In information theory and semiotics, artificial languages represent rule-governed structures whose strict operational definitions eliminate ambiguity, ensuring predictable syntactic analysis and deterministic semantic comprehension across diverse technological or theoretical paradigms.
Ultimately, artificial languages operate at the critical intersection of human agency and systematic formalization. Whether conceived to break international geopolitical barriers, test hypotheses regarding linguistic relativity, or facilitate seamless human-computer interaction, they embody a rationalistic intervention into the fundamentally evolutionary domain of human communicative behavior.
2. Etymology & Linguistic Origin
The term artificial derives from the Latin artificialis, an adjective originating from artificium, meaning “craftsmanship, trade, or artistic skill,” compounded from ars (stem art-, “skill, method, craft”) and facere (“to make, do, or produce”). The noun language traces back through Old French langage to the Vulgar Latin linguaticum, rooted in the Classical Latin lingua (“tongue, speech, dialect”), which itself descends from the Proto-Indo-European root *dnghu- (“tongue”). Historically, the juxtaposition of these concepts gained prominent traction during the seventeenth-century scientific revolution in Western Europe.
During the Renaissance and Enlightenment, early thinkers described such projects as lingua rationalis, lingua universalis, or “philosophical languages.” Gottfried Wilhelm Leibniz and Francis Bacon referred to programmatic linguistic creation as an attempt to formulate a “real character” or an ars signorum. The term “constructed language” emerged more prominently in twentieth-century descriptive linguistics to avoid the pejorative connotations sometimes associated with “artificial,” which skeptics equated with falseness or structural sterility. The modern colloquial blend conlang was coined in the early 1990s on digital forums such as the CONLANG listserv, establishing a clear professional taxonomy among linguistic hobbyists, typologists, and sociolinguists.
3. Pronunciation & Grammatical Form
The standard pronunciation of “artificial language” in International Phonetic Alphabet (IPA) transcription is /ˌɑːrtɪˈfɪʃəl ˈlæŋɡwɪdʒ/ in General American English and /ˌɑːtɪˈfɪʃəl ˈlæŋɡwɪdʒ/ in Received Pronunciation. Morphologically, the compound consists of the descriptive adjective “artificial” (comprising the root artifice and the adjectival suffix -ial) modifying the countable abstract noun “language.”
Grammatically, “artificial language” behaves as a countable noun phrase (e.g., “Scholars have recorded hundreds of artificial languages throughout modern history”) but can also function generically as an uncountable collective noun representing the medium or discipline (e.g., “The paradigm of artificial language poses intriguing cognitive challenges”). Attributively, the phrase can serve as an adjectival modifier, as in “artificial language learning (ALL) paradigms” or “artificial language acquisition studies.” Accepted taxonomic variants in academic literature include “constructed language,” “invented language,” “planned language,” and “synthetic language,” although “synthetic language” frequently conflicts with the morphological typology describing highly inflectional natural languages.
4. Detailed Conceptual Explanation
To conceptualize an artificial language comprehensively, one must contrast its genesis, development, and teleology against those of natural languages. Natural languages (such as English, Mandarin, or Arabic) emerge via generational transmission, characterized by phonetic reduction, lexical borrowing, grammaticalization, semantic shift, and irregular structural artifacts. In contrast, an artificial language begins with an explicit blueprint. The creator, often termed an artificer or conlanger, defines phonemic inventories, establishes phonotactic constraints, designs morphological paradigms, and delineates syntactic rules prior to communicative execution.
The conceptual scope of artificial languages spans a vast spectrum ranging from highly naturalistic human idioms to rigid non-human computational constructs. Naturalistic constructed languages honor cross-linguistic universals, incorporating phonological asymmetries, suppletion, and idiom to achieve psychological authenticity or historical depth. In contrast, non-naturalistic or axiomatic languages deliberately circumvent or subvert human linguistic tendencies. These formal variants seek absolute semantic fidelity, categorical rigor, or computational tractability, eradicating homonymy, polysemy, and grammatical exceptions.
The functional boundaries of artificial languages are demarcated by their operational goals. An artificial language can be designed for international utility, aesthetic world-building, experimental psychological control, or computational instruction. In psycholinguistics, artificial languages are intentionally stripped of extraneous socio-cultural variables, isolating specific morphosyntactic structures to empirically measure how human infants or adults acquire rule-based symbolic behaviors. In formal semantics and analytic philosophy, artificial languages function as idealized mirrors designed to clarify the truth conditions of propositions without the obfuscating vagaries of colloquial rhetoric.
Furthermore, artificial languages challenge conventional notions regarding the relationship between thought and grammar. By constructing an operational medium from first principles, linguistic engineers can intentionally amplify, compress, or redistribute semantic features. This conscious intervention reveals whether human cognition easily accommodates novel linguistic categories or reflexively bends novel grammars back toward standard human typological distributions.
5. Historical Development
The documented history of artificial language development extends across centuries, reflecting evolving philosophical paradigms, geopolitical crises, and technological breakthroughs. One of the earliest documented instances is the Lingua Ignota, devised in the twelfth century by the Christian mystic Hildegard of Bingen. Employing an invented alphabet of twenty-three characters termed the litterae ignotae alongside a hierarchical glossary of mystical and liturgical nouns, Hildegard utilized her system primarily for spiritual contemplation and divine praise.
During the seventeenth century, Western intellectual thought underwent a profound transformation toward systematic rationalism. Philosophers such as Francis Bacon, René Descartes, and John Wilkins sought to construct universal “real characters” and philosophical languages. In 1668, John Wilkins published his landmark work, An Essay Towards a Real Character, and a Philosophical Language, attempting to categorize all existing knowledge into a rational taxonomy of genera, differences, and species, where the very structure of a word would signify its ontological position in the cosmos. Concurrently, Gottfried Wilhelm Leibniz conceptualized the characteristica universalis, an idealized symbolic framework envisioned to convert philosophical dispute into formal mathematical calculation.
By the late nineteenth and early twentieth centuries, industrialization, global migration, and catastrophic geopolitical conflicts shifted the paradigm toward international auxiliary languages (IALs). In 1879, Johann Martin Schleyer introduced Volapük, which captured global attention but quickly fragmented due to structural rigidity. In 1887, L. L. Zamenhof published Unua Libro, introducing Esperanto. Designed with a regularized agglutinative morphology and a romance-based lexicon, Esperanto emerged as the most successful auxiliary language in history, developing an authentic community of native and fluent speakers, original literature, and international cultural institutions.
The mid-twentieth century gave rise to formal logical languages and engineered communication frameworks. In 1955, sociologist James Cooke Brown introduced Loglan to empirically test the Sapir-Whorf hypothesis regarding linguistic relativity, which later spurred the development of Lojban. Simultaneously, the rise of computer science necessitated formal algorithmic languages such as Fortran, Lisp, and Algol, which solidified the theoretical division between naturalistic idioms and formal machine-interpretable systems. The late twentieth and early twenty-first centuries witnessed the dramatic proliferation of artistic languages (artlangs), exemplified by J. R. R. Tolkien’s Elvish tongues (Quenya and Sindarin), Marc Okrand’s Klingon, and David J. Peterson’s Dothraki, establishing artificial language construction as an established literary craft and academic discipline.
6. Theoretical Foundations
The study of artificial languages is informed by multiple theoretical frameworks across formal linguistics, cognitive psychology, and the philosophy of language. At the core of linguistic analysis lies generative grammar, introduced by Noam Chomsky. Generative linguistics posits that the human mind possesses an innate Universal Grammar (UG) that restricts the possible structural configurations of human languages. When applied to artificial languages, generative theory asks whether constructed languages conform to or violate the principles and parameters of UG. Experiments indicate that while human learners can acquire UG-violating artificial rules through conscious mnemonic strategies, the neural pathways activated differ significantly from the natural, subconscious language acquisition mechanisms observed during early developmental stages.
Complementing the formal generative perspective is functional-typological linguistics, represented by scholars such as Joseph Greenberg. Typological frameworks classify languages based on universal cross-linguistic patterns, such as word-order correlations (e.g., Subject-Verb-Object versus Subject-Object-Verb) and implicational hierarchies. Artificial languages engineered without regard for these implicational universals often encounter extreme learning resistance among human subjects, demonstrating that human cognitive processing exhibits deeply ingrained biases toward structural economy, harmonized branching directionality, and communicative efficiency.
From the philosophical perspective, the semiotic theories of Charles Sanders Peirce and Gottlob Frege furnish the epistemological architecture for evaluating artificial systems. Frege’s seminal distinction between sense (Sinn) and reference (Bedeutung) underpins the construction of engineered and logical languages. Natural languages frequently exhibit referential opacity, context-dependent indexicality, and lexical ambiguity. Philosophical artificial languages strive to attain referential transparency, establishing an invariant, bijective correspondence between symbolic expressions and their semantic extensions, thereby eliminating the communicative distortions inherent in vernacular speech.
7. Key Components, Types & Dimensions
Artificial languages can be systematically classified according to their source material, their structural architecture, and their underlying socio-functional objectives. The primary taxonomic dimensions include:
- A Priori Languages: Systems whose vocabulary, grammar, and phonology are invented entirely from scratch or derived from philosophical taxonomies, exhibiting no genetic relationship to pre-existing natural languages (e.g., Wilkins’ Philosophical Language, Solresol, Ro).
- A Posteriori Languages: Systems whose lexical elements, morphological paradigms, and syntactic conventions are systematically drawn from, synthesized, or modeled upon existing natural language families (e.g., Esperanto, Interlingua, Latino sine flexione).
- International Auxiliary Languages (IALs): Pragmatic linguistic tools engineered to foster equitable, cross-border communication among speakers lacking a shared native tongue, emphasizing maximal regularity, phonetic accessibility, and rapid pedagogical acquisition (e.g., Esperanto, Ido, Occidental).
- Engineered Languages (Englangs): Rigorously designed systems created to fulfill precise philosophical, logical, or scientific criteria, typically subdivided into:
- Logical Languages (Loglangs): Built upon first-order predicate calculus to ensure absolute syntactic unambiguity (e.g., Loglan, Lojban).
- Philosophical Languages: Structured around systematic conceptual ontologies where word composition directly mirrors taxonomical categorization.
- Experimental Languages: Devised to test specific cognitive, psychological, or linguistic hypotheses in controlled empirical environments (e.g., Ithkuil, designed to maximize morphosemantic density and eliminate cognitive ambiguity).
- Artistic Languages (Artlangs): Systems constructed for creative satisfaction, aesthetic experimentation, and literary or narrative world-building, comprising:
- Fictional Languages: Designed to evoke cultural, psychological, and historical authenticity within narrative worlds (e.g., Tolkien’s Quenya and Sindarin, Okrand’s Klingon, Peterson’s Valyrian).
- Micronational Languages: Created by micro-communities or sovereign micronational projects for cultural identity (e.g., Talossan).
- Personal Languages (Autolangs): Invented for private introspection, personal diary encryption, or individual artistic enjoyment.
- Formal and Computational Languages: Non-naturalistic symbolic systems characterized by explicit alphabet sets, production rules, and operational semantics designed for algorithmic execution, automated theorem proving, and machine computation (e.g., Python, C++, SQL, regular expressions).
8. Examples & Illustrative Cases
Examining concrete implementations clarifies the vast divergence in structural paradigms among artificial languages:
Case 1: Esperanto (The Auxiliary Prototype). Created by L. L. Zamenhof, Esperanto features sixteen invariant grammatical rules without irregularities. Its morphology is strictly agglutinative, operating on invariant roots combined with transparent affixes. For example, the root san- denotes health: sana (healthy), sano (health), sani (to be healthy), sanigi (to cure/make healthy), and malsano (illness, using the antonym prefix mal-). Grammatical cases are reduced to the nominative and accusative (marked by -n), providing flexible syntactic word order while maintaining grammatical precision.
Case 2: Lojban (The Logical Engine). Developed by the Logical Language Group, Lojban eliminates syntactic ambiguity through an unambiguous context-free grammar based on predicate calculus. Its predicates, known as selbri, govern specific numbers of arguments (sumti). The sentence “The dog sees the cat” is cast as a predicate relationship where the verb acts as an invariant logical relation linking relational arguments, preventing structural polysemy and parsing ambiguities common to human natural language discourse.
Case 3: Klingon (The Typological Counter-Model). Commissioned for the Star Trek media franchise and designed by linguist Marc Okrand, Klingon (tlhIngan Hol) deliberately violates the most common typological defaults of human languages to evoke an alien aesthetic. Okrand selected an Object-Verb-Subject (OVS) basic word order—the rarest word order among Earth languages, observed in fewer than one percent of natural tongues—and introduced harsh phonemic contrasts, such as voiceless velar lateral affricates /tɬ/ alongside retroflex consonants, while omitting common human sounds such as the simple voiceless bilabial plosive /p/ in standard registers.
9. Measurement & Assessment
Assessing the efficacy, structural stability, and cognitive processing of an artificial language requires distinct qualitative and quantitative metrics, depending on whether the system is an auxiliary, engineered, or experimental construct.
In cognitive and psycholinguistic research, artificial language learning paradigms utilize formal metrics such as grammaticality judgment tasks (GJTs), reaction times (RT), and event-related brain potentials (ERPs). During ERP testing, researchers monitor specific electrophysiological signatures, such as the P600 component—an index of syntactic violation and structural repair—and the N400 component, which reflects lexical-semantic integration difficulty. By exposing participants to micro-artificial grammars, investigators isolate the cognitive workload associated with learning specific syntactic parameters (such as head-initial versus head-final branching) under strictly monitored conditions.
For international auxiliary languages, assessment shifts toward communicative efficiency, morphological transparency, and lexical learnability. Quantitative evaluation involves measuring root-morpheme ratios, polysemy indexes, inflectional irregularity counts, and phonemic inventory overlap against major world language families. Typological distance metrics determine the systemic bias inherent in an IAL, calculating whether structural parameters disproportionately favor specific natural language groups (such as Indo-European over Sino-Tibetan or Niger-Congo). In formal computational systems, evaluation centers on parsing complexity, algorithmic decidability, context-free determinism, and execution latency.
10. Applications & Practical Significance
Artificial languages fulfill vital roles across diverse disciplines, demonstrating that linguistic engineering yields far-reaching intellectual, clinical, and industrial utilities.
In neuroscience and developmental psychology, artificial languages serve as the gold standard for investigating language acquisition mechanisms. Natural languages are confounded by uncontrolled variables, including prior semantic associations, idiosyncratic cultural memories, and socio-economic biases. By utilizing miniature artificial languages, developmental researchers can introduce infants and adult learners to tightly controlled auditory streams, precisely evaluating statistical learning capabilities, transitional probability tracking, and rule extraction without the contaminating noise of preexisting linguistic exposure.
In education and pedagogy, learning a regularized artificial language like Esperanto produces documented propaedeutic effects. Empirical studies indicate that students who study Esperanto prior to learning a natural second language (such as French, German, or English) acquire subsequent languages faster and with higher grammatical comprehension than peers who devote that equivalent time exclusively to the target natural tongue. Esperanto’s transparent morphology provides a metacognitive scaffold, enabling learners to internalize structural linguistic abstractions (such as case, tense, aspect, and derivation) without being hindered by irregular conjugations or orthographic inconsistencies.
In entertainment, digital gaming, and world-building industries, artificial languages enhance dramatic realism and cross-cultural immersion. Media franchises routinely hire professional linguists to construct fully functional languages with complete grammars, historical sound changes, and colloquial registers. Concurrently, in computational linguistics and artificial intelligence, the formal study of artificial grammars provides the baseline for natural language processing (NLP), syntax parsing trees, prompt engineering architectures, and intermediate representations in machine translation systems.
11. Research & Empirical Evidence
Substantial empirical research has confirmed the validity of artificial languages as investigative instruments. Landmark work by cognitive psychologists Jeffrey Elman, Jenny Saffran, Richard Aslin, and Elissa Newport demonstrated that eight-month-old human infants can rapidly discover word boundaries in a continuous stream of an artificial language solely through statistical learning. Saffran and colleagues exposed infants to continuous synthetic speech comprising three-syllable nonsense words generated within an artificial grammar. The infants used the statistical transitional probabilities between syllables to differentiate recurring word units from random syllable sequences across word boundaries, fundamentally altering modern theories of speech perception.
In clinical neuroimaging, research led by Angela Friederici and colleagues investigated brain activation during the acquisition of miniature artificial grammars. Utilizing functional Magnetic Resonance Imaging (fMRI), the researchers demonstrated that artificial rules adhering to the universal structural principles of natural languages recruit the classic linguistic network, specifically Broca’s area (Brodmann Area 44/45) and Wernicke’s area. Conversely, when subjects were exposed to non-structure-dependent rules (such as placing a negative marker strictly at a fixed serial position like the fourth word, regardless of syntactic constituency), Broca’s area failed to activate systematically, and subjects relied instead on general working-memory and visuospatial computational networks in the prefrontal cortex.
Sociolinguistic field research has also examined organic community formation around artificial auxiliary systems. Extensive fieldwork conducted by Humphrey Tonkin and Mark Fettes documented authentic intergenerational transmission among native Esperanto speakers (denaskuloj). These studies revealed that children exposed to Esperanto from birth naturally regularize any lingering systemic anomalies and spontaneously create new idioms, validating the hypothesis that human cognitive faculties inherently impose organic evolutionary dynamics onto engineered systems once communicative usage becomes communal and generational.
12. Cultural & Cross-Cultural Considerations
Although artificial languages are engineered to transcend historical and regional contingencies, they remain inextricably tied to cultural paradigms. International auxiliary languages developed during the late nineteenth and early twentieth centuries claimed global neutrality, yet their phonological repertoires, syntactic frameworks, and vocabularies were overwhelmingly Eurocentric. Systems like Esperanto, Volapük, and Interlingua derived their roots almost exclusively from Romance, Germanic, and Slavic languages, imposing a structural burden on speakers of non-Indo-European languages (such as Japanese, Swahili, or Arabic), who must memorize unfamiliar root inventories and grammatical distinctions like grammatical gender or accusative marking.
Conversely, non-Western linguistic traditions have explored constructed language design from distinct philosophical viewpoints. In Islamic intellectual history, the sixteenth-century polymath Muhyi-i Gülşeni devised Bâleybelen, an artificial language constructed to transmit Sufi mystical knowledge, utilizing Arabic, Persian, and Ottoman Turkish lexical and morphological models. In contemporary cross-cultural contexts, modern artificial language projects prioritize typological diversity, incorporating non-linear morphology, ergative-absolutive alignment, tonal distinctions, and phonemic click consonants to dismantle Western grammatical hegemony.
Moreover, the adoption of an artificial language reflects cultural identity dynamics. While critics often dismiss planned languages as lacking soul or cultural heritage, active speaker communities generate distinct cultural artifacts, including original poetry, folk traditions, theatrical productions, and sociolects. The socio-cultural cohesion within artificial language networks illustrates that culture is not solely an ancestral inheritance; it can coalesce rapidly around shared symbolic systems and mutual ideological commitments.
13. Criticisms, Debates & Limitations
The concept and deployment of artificial languages have provoked persistent academic, philosophical, and social debate. A primary critique, historically advanced by the French linguist Antoine Meillet, posits that a living language cannot be disconnected from the organic social milieu of its historical community. Meillet argued that artificial auxiliary languages lack historical vitality, cultural resonance, and unconscious idiomatic richness, leaving them functionally sterile and incapable of capturing the emotional nuances of human literature and oral tradition.
A related controversy concerns the inevitable tension between planning and natural evolution. Linguistic engineers often aim for structural immutability, establishing canonical grammars to prevent dialectal fragmentation. However, sociolinguistic evidence demonstrates that whenever an artificial language attains a living speaker base, standard sociolinguistic processes—such as phonetic drift, slang generation, dialectal variation, and semantic specialization—inevitably emerge. Creators are confronted with an irreconcilable paradox: to remain entirely pure and uniform, the language must remain unpopulated; once widely spoken, it will inexorably break free from central planning and evolve organically, mirroring the very natural languages it sought to refine.
In cognitive and philosophical domains, debates persist regarding the linguistic relativity hypothesis. While engineered languages like Lojban and Ithkuil were constructed to test whether formal grammars can eliminate illogical thought patterns or accelerate human cognitive processing, empirical confirmation remains sparse. Skeptics point out that human cognition frequently operates via analogical, pragmatic, and heuristic processes rather than pure symbolic logic, meaning that forcing thought through ultra-dense or purely algorithmic linguistic channels produces mental fatigue rather than heightened conceptual clarity.
14. Related Terms & Distinctions
To prevent taxonomic confusion, it is necessary to contrast “artificial language” with structurally adjacent concepts:
- Natural Language: An organically evolved human communication system (e.g., Japanese, Spanish, Navajo) that emerges spontaneously within a speech community across generations. In contrast, an artificial language is intentionally authored prior to communicative deployment.
- Pidgin: A simplified, contact-induced language that develops spontaneously between two or more groups lacking a shared language, possessing no native speakers. Unlike artificial languages, pidgins are unplanned, evolving ad hoc out of immediate economic or social necessity.
- Creole: A stable, fully developed natural language that evolves from a pidgin through the process of nativization by child learners. Creolization represents natural, unconscious grammaticalization, whereas artificial languages emerge from conscious intellectual design.
- Formal Language: A purely mathematical set of finite strings composed of symbols governed by rigid production rules (e.g., propositional calculus, lambda calculus). While many artificial languages are formal, others (such as fictional or auxiliary languages) are designed for vocalized human communication and permit expressive pragmatic flexibility.
- Argot or Cant: A specialized, often secretive vocabulary utilized by a subculture or socio-professional group to prevent outsiders from understanding their discourse. Argots modify the lexical layer of a pre-existing natural language rather than inventing an independent phonology, morphology, and syntax.
- Constructed Script (Neography): An invented writing system (e.g., Tolkien’s Tengwar, the Klingon pIqaD). A constructed script represents visual graphemic engineering, whereas an artificial language involves the underlying systemic design of phonology, grammar, and vocabulary.
15. Summary / Key Takeaways
Artificial languages represent the conscious, deliberate engineering of communicative systems, operating in sharp contrast to the slow, evolutionary processes of natural human tongues. Spanning a functional continuum from universal auxiliary languages (Esperanto) to strict logical systems (Lojban) and narrative artistic languages (Klingon, Elvish), artificial languages embody centuries of intellectual endeavor aimed at perfecting, exploring, or celebrating the mechanics of human speech.
Within academic research, artificial languages serve as essential instruments for investigating cognitive architecture, developmental statistical learning, and universal syntactic boundaries. They eliminate confounding sociolinguistic variables, allowing neuroscientists and psychologists to isolate linguistic variables under laboratory conditions. Concurrently, their historical trajectory illustrates the inescapable power of human communicative evolution: whenever an artificial system transitions from a theoretical artifact to an active social medium, it begins to assimilate the very characteristics—phonetic erosion, semantic innovation, and dialectal differentiation—that define natural language life.
Ultimately, artificial languages illuminate the nature of human communicative competence. By constructing languages from first principles, human beings do not merely create new idioms; they dissect the boundaries of their own minds, demonstrating how deeply rule-bound, flexible, and fundamentally creative human linguistic expression remains.
References
- Chomsky, N. (1965). Aspects of the theory of syntax. MIT Press. https://mitpress.mit.edu/9780262527408/aspects-of-the-theory-of-syntax/
- Eco, U. (1995). The search for the perfect language (J. Fentress, Trans.). Blackwell. https://en.wikipedia.org/wiki/The_Search_for_the_Perfect_Language
- Friederici, A. D., Bahlmann, J., Heim, S., Schubotz, R. I., & Anwander, A. (2006). The brain differentiates human and non-human grammars: Functional localization and structural connectivity. Proceedings of the National Academy of Sciences, 103(7), 2458–2463. https://doi.org/10.1073/pnas.0509389103
- Okrent, A. (2009). In the land of invented languages: Esperanto rock stars, Klingon poets, Loglan lovers, and the mad dreamers who tried to build a better language. Spiegel & Grau. https://en.wikipedia.org/wiki/In_the_Land_of_Invented_Languages
- Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274(5294), 1926–1928. https://doi.org/10.1126/science.274.5294.1926