Social Communication

Treat the relational-pragmatic layer of language — turn-taking, relevance, politeness, repair, register, common ground — as a layer distinct from grammatical correctness that can fail on its own, so a child with clean syntax may still be unable to hold a conversation.

Core Idea

Social communication, in the speech-language-pathology and clinical-pragmatics sense, names the relational-pragmatic layer of language use — the norm-governed rules and expectations that determine how participants take turns, stay relevant, manage face and politeness, repair breakdowns, adjust register, and sustain relationships through the exchange itself — held distinct from the phonological, morphological, syntactic, and lexical correctness of the utterances. The structural commitment is a two-layer model of any communicative act: a propositional layer (what was said, its truth-conditional content) and a relational-pragmatic layer (how it was said and how it positioned participants relative to each other, the conversational topic, and the relationship). The two layers can fail independently. A child whose phonology, syntax, and vocabulary are fully intact may nonetheless struggle with turn-taking (interrupting, monopolising the floor), relevance (introducing off-topic content without bridging), politeness (using forms too blunt or too distant for the relational context), or repair (missing cues that the listener has not understood and failing to rephrase). That profile of difficulty with the relational layer while the propositional layer is unimpaired is social communication disorder (SCD), a DSM-5 diagnosis distinct from autism spectrum disorder, and it is the clinical category social communication names.

The theoretical scaffolding that makes the relational-pragmatic layer tractable in clinical and assessment practice draws on several converging frameworks: Grice's cooperative maxims (quantity, quality, relation, manner) specify the implicatures a speaker is expected to generate and the violations that produce pragmatic failures; Sacks, Schegloff, and Jefferson's conversation-analytic turn-taking model specifies the adjacency-pair structure and transition-relevance-point mechanisms that govern who speaks when; Brown and Levinson's politeness theory specifies the face-threatening-act calculus and the positive- and negative-politeness strategies speakers deploy to manage it; and common-ground maintenance — tracking what speaker and listener mutually know and have established across the conversation — is the running inferential task that successful social communication requires. In clinical practice, these frameworks ground the assessment of pragmatic difficulty (the Social Communication Questionnaire, the ADOS social-communication domain, the Pragmatics Profile) and the design of social-communication intervention: social stories, video modelling, scripted exchange practice, and explicit instruction in listener-cue interpretation all target the relational-pragmatic layer directly without modifying the phonological or grammatical architecture.

Structural Signature

Sig role-phrases:

• the propositional layer — what was said: the truth-conditional content carried by intact phonology, syntax, and lexicon
• the relational-pragmatic layer — how it was said and how it positioned participants: the norm-governed, face-managing, relationship-sustaining function of the exchange
• the layer-independence — the structural commitment that the two layers can succeed or fail separately, so the propositional layer can test clean while the relational layer is impaired
• the pragmatic subsystems — the named, theory-grounded parts of the relational layer: turn-taking (adjacency-pair / transition-relevance), relevance and the cooperative maxims, politeness and face management, repair, register adjustment, and common-ground bookkeeping
• the layer-cut diagnostic — the binary first move localizing a conversational misfire to the relational vs. propositional layer, making visible a profile correctness-based assessment is blind to
• the difference-versus-deficit boundary — the check, posable only because the layer is norm-governed, distinguishing a culturally variable pragmatic pattern from a genuine impairment
• the SCD-versus-ASD boundary — the check on whether isolated relational impairment stands alone or travels with restricted, repetitive behaviour, deciding diagnosis and pathway
• the treat-the-relational-layer prescription — the intervention rule that targets the impaired subsystem directly (social stories, video modelling, scripted exchange) and leaves the intact architecture alone

What It Is Not

• Not communication in general. Social communication is the relational-pragmatic layer specifically — turn-taking, relevance, face, repair, register, common ground — held distinct from the propositional layer of what is said. It names how an exchange positions participants and sustains a relationship, not the whole act of conveying content; the propositional layer is a separate object that can test entirely clean while this layer fails.
• Not a grammar, vocabulary, or syntax problem. The defining clinical profile is intact phonology, syntax, and lexicon alongside conversational breakdown. A correctness-based assessment finds nothing wrong at the propositional layer and is structurally blind to the difficulty — which is precisely why social communication disorder needs its own category and its own assessment.
• Not autism spectrum disorder. Social communication disorder is a DSM-5 diagnosis distinct from ASD: it is isolated relational-pragmatic impairment that does not travel with restricted, repetitive behaviour. Reading every pragmatic difficulty as autism collapses the SCD-versus-ASD boundary that determines which diagnosis and which intervention pathway applies.
• Not every unusual pragmatic pattern is a deficit. Pragmatic norms are culturally variable, so a non-native speaker's politeness or turn-taking pattern is a difference, not an impairment. Only a framework that treats the relational layer as norm-governed can pose the difference-versus-deficit question; scoring a violation of one's own community norms as a disorder mistakes cultural variation for pathology.
• Not a single skill. What the construct names is a bundle of named, theory-grounded subsystems — turn-taking, the cooperative maxims, politeness/face management, repair, register, common-ground bookkeeping — that are separately observable and can fail one or two at a time. Treating it as one undifferentiated competence loses the within-layer localization an assessment depends on.
• Not generic "social skills" or workplace soft skills. Those broader trainings borrow the component primes (social norms, face management, turn-taking) applied to new substrates, but social communication as used here is the clinical bundling of those components into a diagnostic category, anchored to a patient, an intact-propositional baseline, and assessment instruments. Strip that clinical apparatus and what travels is the components, not this construct.

Scope of Application

Social communication lives across the clinical-pragmatics subfields of speech-language pathology and adjacent communication-disorder fields; its reach is within that domain — the workplace soft-skills, online-moderation, and dialogue-system uses are the component primes (social_norms, common_ground, face_management, turn-taking, repair) applied elsewhere, not the clinical bundle travelling.

• Paediatric language assessment — the home: diagnoses social communication disorder (SCD) as a DSM-5 category of isolated relational-pragmatic impairment, distinct from autism, with instruments like the Pragmatics Profile.
• Autism-spectrum assessment — reads the social-communication domain (Social Communication Questionnaire, ADOS social-communication domain, SCERTS) as a core diagnostic feature.
• Aphasiology and right-hemisphere-damage work — assesses acquired pragmatic deficits (flattened prosody, lost sarcasm comprehension, topic-maintenance failure) against the same two-layer framework.
• Augmentative and alternative communication (AAC) — designs devices and training around relational competence — turn-taking, repair, register — not message production alone.

Clarity

Naming social communication as a layer in its own right makes visible a child who would otherwise be missed entirely: the one whose phonology, syntax, and vocabulary all test clean, yet who cannot hold a conversation. Without the construct, that profile is invisible to a language assessment built around utterance correctness — the clinician finds nothing wrong at the propositional layer and has no place to file the difficulty. By separating can the utterance be formed correctly? from can it be deployed appropriately in a relational context?, the framework gives the relational failures — interrupting, drifting off-topic, blunt register, missed repair cues — a layer of their own, with their own assessment and their own intervention. This is the clarifying move that licenses social communication disorder as a category: a real, treatable difficulty that grammar-and-vocabulary testing is structurally blind to.

The construct also sharpens several distinctions that clinical practice cannot afford to blur. It separates a pragmatic violation (an off-topic turn, a flat prosody, a face-threatening bluntness) from a grammatical error, so the two carry independent diagnostic weight rather than being lumped as generic "language problems." It forces the difference-versus-deficit question, since pragmatic norms are culturally variable — a non-native speaker's politeness pattern is a cultural difference, not an impairment, and only a framework that treats the relational layer as norm-governed can even pose the distinction. And it draws the clinically critical line between social communication disorder and autism spectrum disorder: by isolating the relational-pragmatic layer as the locus of difficulty, it lets the diagnostician ask whether pragmatic impairment stands alone or travels with restricted, repetitive behaviour — a boundary that determines which diagnosis, and which intervention pathway, applies. In each case the sharpened question is the same in form: localise the breakdown to a layer, and treat that layer directly rather than retraining an architecture that is already intact.

Manages Complexity

A conversation that goes wrong can go wrong in indefinitely many ways — a child interrupts, drifts off-topic, stands too close, speaks too bluntly, misses that the listener is lost, fails to soften a request, talks past the other's knowledge — and each surface failure has its own descriptive vocabulary scattered across Grice, conversation analysis, politeness theory, and common-ground research. Social communication compresses that sprawl in two strokes. First, it collapses the whole catalogue of conversational misfires onto a single layer distinct from utterance correctness: every one of them is a failure of the relational-pragmatic layer rather than the propositional layer, so the clinician's first move is not to enumerate the dozen things that can go wrong but to ask one binary — is the breakdown propositional (phonology, syntax, lexicon) or relational? That single cut tells the clinician where to look and, decisively, makes visible the child who tests clean on every grammar-and-vocabulary measure yet cannot hold a conversation, a profile a correctness-based assessment is structurally blind to. The qualitative diagnosis reads off the layer: intact propositional layer plus impaired relational layer is social communication disorder, and the intervention follows — treat the relational layer directly (social stories, video modelling, scripted exchange, listener-cue instruction) and leave the already-intact architecture alone.

Second, the construct organises the relational layer itself, which would otherwise be an open list of "soft" expectations, into a small set of named, theory-grounded subsystems an assessor can track one at a time: turn-taking (the adjacency-pair and transition-relevance machinery of who speaks when), relevance and the cooperative maxims (the implicature-governing expectations whose violation produces off-topic or under-/over-informative turns), politeness and face management (the face-threatening-act calculus and its positive/negative strategies), repair (the routine for detecting and fixing a comprehension breakdown), register adjustment, and the running common-ground bookkeeping. The full intricacy of "good conversational behaviour" reduces to a checklist of these subsystems, each separately observable, so a Pragmatics Profile or ADOS social-communication domain reads a difficulty off as residing in one or two of them rather than as diffuse social awkwardness. The compression also supplies two branch points the clinician must read but the framework now makes crisp: the difference-versus-deficit branch, since the norms are culturally variable, so a non-native politeness pattern is scored as cultural difference, not impairment — a question only a norm-governed-layer model can pose; and the SCD-versus-ASD branch, since isolating the relational layer lets the diagnostician ask whether pragmatic impairment stands alone or travels with restricted, repetitive behaviour, which determines diagnosis and pathway. The analyst thus tracks one layer-cut, a handful of named subsystems, and two branch conditions, and reads off the diagnosis, the breakdown locus, and the intervention target — replacing an unbounded inventory of conversational failures with a low-dimensional, layered, branch-structured account.

Abstract Reasoning

Social communication licenses reasoning that resolves a communicative failure onto a two-layer model — propositional versus relational-pragmatic — and then localizes within the relational layer to a named subsystem. The defining diagnostic move runs from a conversation that goes wrong to a layer, against the grain of a correctness-based assessment. A child who interrupts, drifts off-topic, stands too close, speaks too bluntly, or misses that the listener is lost is read not as having a "language problem" in general but as failing the relational-pragmatic layer specifically, with the propositional layer (phonology, syntax, lexicon) intact. The surface signature is the conversational misfire; the inferred locus is the relational layer. Reasoning FROM "this child tests clean on grammar and vocabulary yet cannot hold a conversation" TO "the breakdown is relational, not propositional" is what makes visible a profile a correctness-based assessment is structurally blind to, and it is the move that licenses social communication disorder as a category: intact propositional layer plus impaired relational layer is SCD.

A within-layer localization move then decomposes the relational layer into named, theory-grounded subsystems and reads a difficulty off as residing in one or two of them. Turn-taking (the adjacency-pair and transition-relevance machinery), relevance and the cooperative maxims (whose violation yields off-topic or under-/over-informative turns), politeness and face management (the face-threatening-act calculus), repair (detecting and fixing a comprehension breakdown), register adjustment, and common-ground bookkeeping are each separately observable. Reasoning FROM "what kind of conversational failure is this" TO "which pragmatic subsystem is impaired" converts diffuse social awkwardness into a located deficit a Pragmatics Profile or ADOS social-communication domain can pin down.

The interventionist move follows directly and carries a do-not-treat clause: because the propositional architecture is already intact, therapy targets the relational layer directly — social stories, video modelling, scripted exchange practice, explicit listener-cue instruction — and predicts gains in the impaired subsystems without modifying the phonological or grammatical machinery. Reasoning FROM "the deficit is relational, the grammar is fine" TO "treat the relational layer and leave the architecture alone" is what keeps intervention from retraining a system that is not broken.

Two boundary-drawing moves discipline the diagnosis, and both are posable only because the construct treats the relational layer as norm-governed. The difference-versus-deficit move asks whether an unusual pragmatic pattern is a culturally variable norm rather than an impairment — a non-native speaker's politeness pattern is a cultural difference, not a disorder — so the reasoner draws the line FROM "is this a violation of the listener's own community norms or merely of mine" TO "deficit versus difference." The SCD-versus-ASD move asks whether the isolated relational impairment stands alone or travels with restricted, repetitive behavior, since only the latter shifts the diagnosis toward autism. Reasoning FROM "does pragmatic impairment co-occur with restricted/repetitive patterns" TO "which diagnosis and which intervention pathway applies" is the boundary that the layer-isolation makes available, determining the clinical route rather than leaving it to global impression.

Knowledge Transfer

Within speech-language pathology and the adjacent clinical fields the construct transfers as full mechanism, and the cargo that carries is the whole clinical apparatus: the propositional/relational two-layer model, the named subsystems (turn-taking, relevance/maxims, politeness/face, repair, register, common ground), the difference-versus-deficit and SCD-versus-ASD boundary moves, and the treat-the-relational-layer-directly prescription. Paediatric language assessment uses it to diagnose social communication disorder as a DSM-5 category distinct from autism. Autism-spectrum assessment reads the social-communication domain (SCQ, ADOS, SCERTS) as a core diagnostic feature. Aphasiology and right-hemisphere-damage work assess acquired pragmatic deficits — flattened prosody, lost sarcasm comprehension, topic-maintenance failure — against the same framework. Augmentative and alternative communication designs devices and training around relational competence, not message production alone. Across these the layer-cut, the subsystem checklist, the two branch conditions, and the intervention targets are the same objects; only the population and the etiology change. The transfer is literal because the substrate — human conversational interaction with its norm-governed relational layer — is held fixed.

Beyond the clinic the situation is the shared-abstract-mechanism case in an unusually pure form, because social communication is not one mechanism but a bundle of them. Its components really do recur across substrates: workplace soft-skills and leadership training, online-community moderation and platform design, conversational-AI dialogue systems, and cross-cultural communication all wrestle with turn-taking, relevance, politeness, repair, register, and common ground. But this breadth is not the clinical construct travelling — it is the breadth of the component primes applied to new substrates. Strip the clinical vocabulary (SCD, SCERTS, ADOS, Pragmatics Profile, the difference-versus-deficit triage) and what is left is the conjunction of social_norms specialized to language use, common_ground, the politeness/face_management calculus, speech_act_theory, code_switching for register, and the turn-taking and repair routines — every one of which is independently housed at the substrate-independent level and does the structural work in the new domain. A forum moderator articulating off-topic rules is doing social_norms plus common_ground; a dialogue system implementing back-channels and clarification requests is borrowing turn-taking and repair from conversation analysis; cross-cultural politeness work ports Brown-Levinson with substrate-varying markers. None of these inherits the clinical layer-model as a diagnostic — there is no patient, no intact-propositional-layer baseline to contrast against, no SCD-versus-ASD branch, no assessment battery built to hold grammar constant.

So the cross-domain lesson — the relational, norm-governed layer of an exchange can succeed or fail independently of its propositional content, and is governed by its own rules for turning, relevance, face, and repair — should be carried by the component primes, not by "social communication," which is the field-specific bundling of those primes into a clinical category. Even within the clinic, the part that genuinely travels (the layer-separability insight) travels via communication and social_norms; what stays home-bound is the diagnostic machinery — social communication disorder as a DSM-5 category, the assessment instruments, the difference-versus-deficit and SCD-versus-ASD boundaries, and the do-not-retrain-the-grammar intervention logic. The honest framing is: the components are general and travel as primes; the clinical construct that assembles them is the domain accent that stays in speech-language pathology (see Structural Core vs. Domain Accent).

Examples

Canonical

The defining instance is the DSM-5 category of social communication disorder (SCD), introduced in 2013. Consider a school-age child who scores in the normal range on every standardized measure of articulation, vocabulary, and grammar — she forms complex, well-ordered sentences — yet in conversation she interrupts, launches into a monologue about her own interest without checking whether the listener follows, takes an idiom literally, and fails to soften a blunt request to a teacher. Standardized language testing, built around utterance correctness, finds nothing wrong. The two-layer model explains why: her propositional layer is intact while her relational-pragmatic layer — turn-taking, relevance, register, common ground — is impaired. Grice's cooperative maxims name what she violates (heaping detail on the wrong topic flouts relation and quantity), and because that impairment stands alone, without restricted, repetitive behavior, the diagnosis is SCD rather than autism.

Mapped back: Her clean articulation, vocabulary, and grammar scores are the intact propositional layer; the interrupting, monologuing, literal idiom, and blunt register are failures of the relational-pragmatic layer — a direct display of the layer-independence. That standardized testing sees nothing is the layer-cut diagnostic revealing a profile correctness-based assessment is blind to, and the absence of restricted, repetitive behavior is the SCD-versus-ASD boundary deciding the diagnosis.

Applied / In Practice

In adult neurology the same two-layer model is deployed after right-hemisphere stroke. Patients with right-hemisphere damage frequently retain fluent, grammatically intact speech — clean phonology, syntax, and vocabulary — yet lose command of the relational-pragmatic layer: flattened prosody, missed sarcasm and non-literal meaning, off-topic over-elaboration, and failure to maintain conversational topic or read a listener's cues. A speech-language pathologist assessing such a patient does not re-test grammar, which is spared, but probes the pragmatic subsystems directly — topic maintenance, inference of implied meaning, prosodic interpretation — and targets therapy at those, using cue-interpretation practice and structured conversation. The framework is what tells the clinician that the deficit is relational, that the intact linguistic architecture should be left alone, and where within the relational layer the breakdown sits — the identical layer-cut and subsystem localization used with children, applied to an acquired etiology.

Mapped back: The spared phonology, syntax, and vocabulary are the intact propositional layer; the lost prosody, sarcasm, and topic control are impairments of the relational-pragmatic layer, another case of the layer-independence. Probing topic maintenance, inference, and prosody is the within-layer localization to named pragmatic subsystems, and treating those while leaving grammar untouched is the treat-the-relational-layer prescription.

Structural Tensions

T1: Clean layer separation versus mutual dependence (the two layers interact even as the model holds them apart). The construct's power is the commitment that the propositional and relational layers can fail independently — a child with clean syntax who cannot converse. That separability is what makes the missed profile visible. But in real conversation the layers lean on each other: a relational failure of common-ground tracking degrades what propositional content a listener can even receive, and a subtle propositional weakness (a narrow vocabulary that forces blunt phrasing) can masquerade as a politeness deficit. Treating the layers as cleanly separable is exactly what surfaces SCD, and exactly what can misattribute a relational-looking symptom whose root is propositional, or miss that fixing one layer requires the other. The idealization that makes the category legible is a simplification of a genuinely coupled system. Diagnostic: Is this relational breakdown independent of the propositional layer, or is a propositional limitation surfacing in relational clothing?

T2: Difference-versus-deficit safeguard versus under-diagnosis (the culturally-humble check can also miss a real impairment). Only a norm-governed-layer model can even pose the difference-versus-deficit question, and it rightly stops a clinician from scoring a non-native politeness pattern as pathology. But the safeguard cuts both ways: a genuine relational impairment in a child from a different cultural or linguistic background can be waved off as "just a cultural difference," delaying an intervention the child needs. The same norm-relativity that protects against over-diagnosing the culturally different can rationalize under-diagnosing them, because the clinician's uncertainty about whose norms apply becomes a reason to withhold a verdict. Deciding whose community norms are the reference is itself contested, and the framework poses the question without settling it. Diagnostic: Is the unusual pattern a violation of the listener's own community norms (deficit) or only of the clinician's (difference) — and is "difference" being invoked to protect the child or to avoid a hard call?

T3: SCD-versus-ASD boundary versus a shared underlying continuum (a categorical cut through possibly graded territory). Isolating the relational layer lets the diagnostician ask whether pragmatic impairment stands alone or travels with restricted, repetitive behavior — the boundary that routes diagnosis and pathway. Clinically indispensable, but the boundary presumes SCD and ASD are cleanly separable categories, when the co-occurrence of pragmatic and restricted-repetitive features is often a matter of degree and developmental timing. A child below the RRB threshold today may cross it later; the "isolated" impairment may be an early or mild point on a continuum the categorical cut treats as a different kind. The boundary that determines the intervention pathway can freeze a graded, evolving picture into a binary that later has to be revised. Diagnostic: Are the restricted, repetitive features genuinely absent, or merely sub-threshold on a continuum the categorical SCD/ASD cut will force into one bin?

T4: Treat-the-layer-directly versus mechanism-blind symptom training (targeting the relational surface can drill scripts without competence). The do-not-retrain-the-grammar prescription is elegant: the architecture is intact, so target the relational subsystem directly with social stories, scripted exchange, and video modelling. But teaching to the observable relational behavior risks producing a child who executes the taught script — the rehearsed greeting, the practiced topic-check — without the underlying common-ground inference that generates appropriate behavior in novel exchanges. The intervention that correctly refuses to fix an unbroken grammar can substitute surface compliance for pragmatic competence, improving the assessed subsystem in-clinic while the generative relational skill it stands for does not transfer. The directness that avoids over-treatment can also mistake the trained behavior for the capacity. Diagnostic: Does the intervention build the inference that generates appropriate relational behavior, or only drill the specific scripted behaviors the assessment scores?

T5: Subsystem checklist versus holistic conversational competence (decomposing the relational layer can lose the thing being decomposed). Organizing the relational layer into named, separately observable subsystems — turn-taking, maxims, face, repair, register, common ground — is what converts diffuse social awkwardness into a located deficit a Pragmatics Profile can pin. But conversational competence is substantially the integration of these subsystems in real time; a child can score adequately on each subsystem in isolation and still fail to hold a conversation because the orchestration across them is what is impaired. Reading a difficulty off as "residing in one or two subsystems" can miss a coordination deficit that lives in no single box. The decomposition that makes the layer assessable can dissolve the integrative competence that is the actual target. Diagnostic: Does this child's difficulty localize to specific subsystems, or do the subsystems test adequately in isolation while their real-time integration fails?

T6: Autonomy versus reduction (its own clinical category or a field-specific bundle of parent primes). "Social communication" is a named DSM-5-anchored clinical construct with proprietary apparatus — the SCD category, the assessment batteries (SCQ, ADOS, Pragmatics Profile), the difference-versus-deficit and SCD-versus-ASD triage, the intact-propositional baseline — and that whole apparatus transfers as mechanism across the clinical fields (paediatric assessment, aphasiology, AAC) where the conversational substrate is held fixed. But it is not one mechanism; it is a bundle, and beyond the clinic what actually travels is the component parents it assembles — social_norms specialized to language, common_ground, face_management, speech_act_theory, code_switching for register, and the turn-taking and repair routines. A forum moderator is doing social-norms-plus-common-ground; a dialogue system borrows turn-taking and repair; cross-cultural work ports Brown-Levinson — none inherits the clinical diagnostic. The tension is between a standalone clinical category that earns its own instruments and the recognition that its cross-domain cargo is the component primes it bundles, not the bundle. Diagnostic: Resolve toward the component primes when asking what travels outside the clinic; toward "social communication" when diagnosing an isolated relational-pragmatic impairment against an intact-propositional baseline.

Structural–Framed Character

Social communication sits at the framed-leaning position on the structural–framed spectrum — a named clinical construct that bundles several component primes into a diagnostic category, distinct in that its portable content is not one skeleton but a conjunction of parents. The criteria pull framed. Evaluative_weight is real: as used, the construct is anchored to social communication disorder — a DSM-5 category with impairment judgments, a deficit-versus-difference triage, and an intact-propositional baseline against which failure is scored — so it renders clinical verdicts, not neutral description (the bare layer model is descriptive, but the construct that earns its instruments is diagnostic). Human_practice_bound is high and doubled: the relational-pragmatic layer it names is itself constituted by human conversational practice (turn-taking, face, politeness norms), and the construct is further constituted by clinical practice — a patient, an assessment battery, a diagnostic pathway — dissolving without both. Institutional_origin is pronounced: SCD, SCERTS, ADOS, the SCQ, the Pragmatics Profile, the difference-versus-deficit and SCD-versus-ASD boundaries are all artifacts of a clinical-diagnostic tradition, not structures found in nature. Vocab_travels fails for the bundle: the clinical apparatus does not port, and beyond the clinic what appears is the component primes applied to new substrates, not the diagnostic category. Correspondingly import_vs_recognize is component-borrowing off-domain: a forum moderator does social_norms plus common_ground, a dialogue system borrows turn-taking and repair — none inherits the clinical layer-model as a diagnostic.

The portable content is a layer-separability insight — the relational, norm-governed layer of an exchange can succeed or fail independently of its propositional content, and is governed by its own rules for turning, relevance, face, and repair — but, unusually, this is not a single skeleton the construct instantiates; it is a bundle it assembles from several umbrella primes. What genuinely travels is those component parents: social_norms specialized to language use, common_ground, face_management (the Brown–Levinson politeness calculus), speech_act_theory, code_switching for register, plus the turn-taking and repair routines from conversation analysis — each independently housed at the substrate-independent level and each doing the structural work in any new domain. The cross-domain reach belongs to those components (and the bare layer-separability insight travels via communication and social_norms), while everything that makes "social communication" the specific named construct — the SCD category, the assessment batteries, the difference-versus-deficit and SCD-versus-ASD triage, the do-not-retrain-the-grammar logic — stays home in speech-language pathology. Its character: an evaluatively-laden, doubly practice-bound, clinically-originated diagnostic construct whose substrate-spanning content is not one prime but the conjunction of component primes it bundles (social_norms, common_ground, face_management, speech_act_theory, code_switching, turn-taking, repair), the clinical bundling itself being the domain accent that does not travel.

Structural Core vs. Domain Accent

This section decides why social communication is a domain-specific abstraction and not a prime, and it carries the case for its domain-specificity — there is no separate section for it. It is an unusual case: the portable content is not one skeleton but a bundle of several parents.

What is skeletal (could lift toward a cross-domain prime). Strip the clinical apparatus and a portable insight survives: the relational, norm-governed layer of an exchange can succeed or fail independently of its propositional content, and is governed by its own rules for turning, relevance, face, and repair. But — and this is what marks social communication out — that content is not a single relational structure the construct instantiates; it is a conjunction the construct assembles from several umbrella primes, each independently portable: social_norms specialized to language use, common_ground, face_management (the politeness/face calculus), speech_act_theory, code_switching for register, plus the turn-taking and repair routines of conversation analysis. Each of these does the structural work in any new substrate on its own; the bare layer-separability insight itself travels via communication and social_norms. That every component is genuinely substrate-portable is exactly why the construct's reach looks so broad — but the breadth belongs to the parts, not to the bundle.

What is domain-bound. Almost everything that makes the concept social communication in particular is speech-language-pathology furniture and none of it survives extraction intact: the two-layer clinical model anchored to an intact-propositional baseline; the diagnostic category of social communication disorder as a DSM-5 entry; the assessment batteries (the Social Communication Questionnaire, the ADOS social-communication domain, SCERTS, the Pragmatics Profile); the difference-versus-deficit triage that only a norm-governed-layer model can pose; the SCD-versus-ASD boundary that routes diagnosis and pathway; and the treat-the-relational-layer-directly, do-not-retrain-the-grammar intervention logic. These are the instruments and the worked cases — the child with clean syntax who cannot converse, the right-hemisphere-stroke patient who has lost sarcasm and prosody — that the clinic actually uses. The decisive test: remove the patient, the intact-propositional baseline, and the diagnostic pathway, and the bundle stops being this construct and reverts to its loose components. A forum moderator articulating off-topic rules is doing social_norms plus common_ground; a dialogue system implementing back-channels and clarification requests is borrowing turn-taking and repair — neither inherits the clinical layer-model as a diagnostic, because there is no grammar being held constant and nothing being scored against a baseline.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose cross-domain transfer is recognition of the same mechanism, not analogy or component-borrowing. Social communication's transfer is bimodal, and doubly so because it is a bundle. Within speech-language pathology and its adjacent clinical fields the whole apparatus travels intact — the layer-cut, the subsystem checklist, the two branch conditions, and the intervention targets are the same objects across paediatric assessment, autism-spectrum assessment, aphasiology, and AAC, because the substrate (human conversational interaction with its norm-governed relational layer) is held fixed. Beyond the clinic the named construct does not travel; what appears is its component primes applied to new substrates — workplace soft-skills training, online-community moderation, conversational-AI dialogue systems, cross-cultural politeness work — each doing social_norms, common_ground, face_management, speech_act_theory, code_switching, turn-taking, or repair, but none inheriting the diagnostic bundle. And when the bare structural lesson is needed cross-domain — that a norm-governed relational layer can fail independently of propositional content — it is already carried, in more general form, by those component parents (and by communication and social_norms). The cross-domain reach belongs to the components; "social communication," as named, is the field-specific bundling of them into a clinical category, and the bundling is the domain accent that stays home.

Relationships to Other Abstractions

Current abstraction Social Communication Domain-specific

Parents (8) — more general patterns this builds on

• Social Communication is a decomposition of Common Ground Prime

  Social communication tracks what interlocutors mutually recognize so deixis, inference, topic maintenance, and appropriately abbreviated turns can work.

• Social Communication is a decomposition of Communication Repair Prime

  Detecting misunderstanding, initiating clarification, restoring alignment, and resuming are a named clinical subsystem of conversational competence.

• Social Communication is a decomposition of Cooperative Principle and Gricean Maxims Prime

  Relevance, informativeness, truthfulness, clarity, and implicature form the norm-governed inferential subsystem behind topic and nonliteral-meaning competence.

• Social Communication is a decomposition of Pragmatic Politeness Strategies Prime

  Politeness and face-management form a named subsystem of the relational layer, modulating face threats through directness, hedging, redress, and restraint.

• Social Communication is a decomposition of Register (Style) Shifting Prime

  Social communication includes selecting formality, technicality, intimacy, and deference appropriate to audience, setting, topic, and task.

• Social Communication is a decomposition of Social Norms Prime

  The relational-pragmatic layer is governed by community expectations for relevance, turn allocation, register, face, topic, and repair.

• Social Communication is a decomposition of Speech Act Theory (Illocution, Perlocution) Prime

  The layer distinguishes literal form from the request, promise, greeting, refusal, implication, or effect an utterance performs in context.

• Social Communication is a decomposition of Turn Taking Prime

  Turn allocation, yielding, overlap management, and transition timing are a separately observable pragmatic subsystem assessed apart from grammar.

Hierarchy paths (26) — routes to 14 parentless roots

• Social Communication → Common Ground → Common Knowledge → Hierarchy → Network → Reservoir-Flux Network → Conservation Laws → Invariance

Show alternative paths (25)

Not to Be Confused With

• Structural language disorder (phonology / syntax / vocabulary). Impairment of the propositional layer — how utterances are formed. Social communication names the relational-pragmatic layer, which can fail while the propositional layer tests entirely clean. Tell: does the child mis-form words and sentences (structural language disorder) or form them correctly yet mis-deploy them in conversation — interrupting, drifting off-topic, blunt register (social communication)? The layer-cut is the diagnostic: propositional versus relational.
• Social (pragmatic) communication disorder (SCD). The DSM-5 diagnostic category — an isolated impairment of the relational-pragmatic layer. Social communication is the layer/construct itself (present in everyone, impaired in some); SCD is the named disorder of that layer. Tell: are you naming the normal-to-impaired capacity for relational-pragmatic language (social communication) or the specific clinical diagnosis of its isolated impairment (SCD)? The construct is the object; the disorder is a verdict on it.
• Autism spectrum disorder (ASD). A diagnosis in which relational-pragmatic impairment travels with restricted, repetitive behaviour. SCD is precisely the case where the pragmatic impairment stands alone, without RRBs. Tell: does the pragmatic difficulty co-occur with restricted, repetitive patterns and interests (ASD) or stand in isolation (SCD)? Reading every pragmatic difficulty as autism collapses the boundary that routes diagnosis and intervention.
• Pragmatics (the linguistics subfield). The academic study of language-in-use — Grice's maxims, speech acts, deixis, implicature. Social communication clinically bundles pragmatics (plus conversation analysis and politeness theory) into an assessable, treatable diagnostic construct anchored to a patient and a baseline. Tell: is it the descriptive theory of how meaning depends on context (pragmatics) or the clinical operationalization of that theory into subsystems, instruments, and a disorder category (social communication)? One is the science; the other its clinical application.
• Generic "social skills" / soft skills. Broad interpersonal-competence training (eye contact, assertiveness, teamwork) applied to workplaces and everyday life. It borrows the same component primes but lacks the clinical apparatus — patient, intact-propositional baseline, assessment battery, difference-versus-deficit triage. Tell: is there a diagnostic baseline against which relational failure is scored (social communication) or general interpersonal coaching with no clinical contrast (soft skills)? Strip the clinic and only the components remain.
• Theory of mind. The cognitive capacity to attribute mental states to others — a substrate that supports pragmatic competence but is not identical to it. A child can have the mentalizing capacity yet fail the motor of turn-taking or repair, or vice versa. Tell: is the deficit in inferring what others know/intend (theory of mind) or in the conversational machinery of turning, relevance, face, and repair (social communication)? ToM underlies part of the relational layer but does not exhaust it.
• The component primes it bundles (social_norms, common_ground, face_management, speech_act_theory, code_switching, turn-taking, repair). The substrate-neutral parents — each independently portable — that social communication assembles into a clinical category. They are what actually travel to moderation, dialogue systems, and cross-cultural work; "social communication" is the field-specific bundle. Tell: strip the clinical apparatus and what remains — norms, common ground, face, speech acts, register, turn-taking, repair, each doing structural work on its own — is these primes, not "social communication." (Treated fully in a later section.)

Neighborhood in Abstraction Space

Social Communication sits in a crowded region of the domain-specific corpus (12^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Voice, Audience & Social Meaning (16 abstractions)

Nearest neighbors

• Axial Coding — 0.88
• Structuralism — 0.88
• Interpersonal Communication — 0.87
• Media Richness Theory — 0.86
• Loaded Question — 0.86

Computed from structural-signature embeddings · 2026-07-12