The "Inefficient" Dialect Is a Myth

1. The claim and its two versions

"Dialects raise communication costs, so speech should be standardized" sounds practical. It names a friction people feel: a Cantonese-only speaker and a Mandarin-only speaker meeting for the first time need a translator, a phrasebook, or weeks of learning. In a hospital, a court, a university seminar, or a job interview, that friction matters. Such friction exists. The question is how it is described, what it is attributed to, and what policy follows from it.

The claim has a strong version and a weak version. The strong version says dialects should be eliminated and replaced by a standard language. That version does not hold up linguistically. A dialect is not an inefficient version of a standard language. It is another complete system of phonology, vocabulary, grammar, pragmatics, and social meaning. It can handle complex thought, humor, intimacy, local history, and technical vocabulary within its domain. To call it "incomplete" or "backward" is to mistake a political hierarchy for a structural fact about language.

The weak version says that in cross-dialect, cross-regional public settings, a shared common language is useful. That version is defensible. Modern states need a common medium for administration, education, public health, law, and mass communication. The question is not whether a common language should exist. The question is what relationship it should have with dialects. Two things need separating: building a common language for public communication, and treating dialects as errors to be erased.

2. The cost belongs to the boundary, not to the dialect

In a multi-dialect setting, communication can carry extra costs. A Cantonese speaker and a Mandarin speaker who share no other language must find a bridge. That bridge is translation, interpretation, code-switching, or mutual learning. The time and attention spent on the bridge are costs that show up in hours and fatigue. But the cost belongs to the boundary between systems. It does not belong to dialect as such. Any two distinct language systems produce this cost when they meet. A Mandarin speaker meeting an English speaker faces the same kind of boundary. An English speaker meeting a French speaker faces it too. Different systems are different. That is where the cost comes from, not from some supposed inefficiency in dialect.

Eliminating dialects does not eliminate boundaries. It relocates them. Suppose a community once conducted local life in a dialect, and public life is then restricted to a standard language. Some cross-dialect costs may fall. New costs appear: converting local experience into the standard, monitoring one's accent, losing the precise social signal that a dialect particle carries. The boundary does not vanish. It moves from between dialects to between dialect speakers and the standard.

Inside a community, a dialect often lowers communication costs. A group of Sichuanese friends chatting in Sichuanese can use tone, rhythm, local idioms, and shared references with high bandwidth. The intimacy, the timing of a joke, the small particles that soften a request or sharpen a tease—these are hard to reproduce in standard Mandarin. This is not nostalgia. It is a description of how shared linguistic knowledge works. A dialect is a tool for lowering friction among people who share it. The cost it imposes on outsiders is a boundary cost, not evidence that the tool is badly made.

Communities are not uniform. Generations differ. Rural and urban speakers differ. Some topics are easier in the standard language because schooling and media have trained people to discuss them that way. Young speakers code-switch without noticing. Dialects are not always more efficient everywhere. But within their appropriate domain, they carry efficiency and pragmatic precision that a standard language cannot automatically replace. Treating the cross-community cost of a dialect as the dialect's own cost is an attribution error.

3. Ambiguity is a working property of language

The standardization argument often assumes that ambiguity is an enemy. If only everyone spoke one precise standard, misunderstanding would shrink. This picture gets language wrong. Natural language works with finite symbols under infinite pressure. Words carry multiple potential meanings. Context, shared knowledge, and pragmatic inference resolve most of them. That arrangement is not a bug. It is what lets language handle new situations, metaphor, humor, and shifts in stance. A language with one exact meaning per word would be brittle.

Standard languages are not exempt. Legal texts are written in standard language and still require judicial interpretation. Scientific papers are written in standard language and still require peer review to clarify claims. Ambiguity is a property of natural language, not a defect specific to dialects. To make "eliminating ambiguity" the reason for standardizing speech is to treat a basic working property of language as a fault to be repaired.

What is often called "dialect ambiguity" is usually outsider ambiguity. Two locals use an expression that puzzles a visitor. Between the locals, the expression is clear because they share context, prosody, and local usage. The dialect does not manufacture ambiguity. Cross-dialect exchange lacks shared context and shared pragmatic habits. Eliminating the dialect does not remove outsider ambiguity. It changes who the outsider is. The visitor who once did not understand the local dialect may become the local speaker who does not command the standard.

Ambiguity is not the same as non-comprehension. Ambiguity is multiple possible interpretations inside a system, often resolved by context. Non-comprehension across dialects is a lack of shared knowledge, sound correspondences, and usage conventions. "I do not understand Cantonese" is not the same as "Cantonese is ambiguous." Mixing the two turns a boundary problem into a supposed defect of the dialect.

4. The historical price of standardization

Language standardization has a history, and the price is often undercounted. After the French Revolution, policies promoted standard French and suppressed regional languages such as Breton, Occitan, and Corsican. The process was tied to the idea that local speech was backward. Generations of speakers learned to feel shame about their first language. In China, the modern national language movement similarly pushed dialects out of many public domains. Systematic dialect protection has appeared only more recently. These histories differ in detail, but they share a pattern: when a language is labeled "nonstandard," its speakers can internalize the label.

That internalization is a communication cost. A speaker who fears that her mother tongue is "not good enough" may hesitate in public, monitor her grammar, flatten her accent, or avoid speaking altogether. The attention spent on self-monitoring is attention not spent on the message. This cost is harder to count than a translator's fee, but it is real. Standardization may reduce some cross-dialect costs while converting others into psychological and cognitive burdens for dialect speakers.

The history also warns against a false choice. A common language can serve public communication without requiring forced assimilation. Standard languages have helped build schools, administrate states, and widen communication across regions. That record does not justify humiliating local speech. The target of criticism should be forced homogenization, language shame, and the closing of public domains to dialects. Not the existence of a common language. A country can need a shared medium for parliament and still allow a grandmother to tell a story in the language she dreams in.

5. Standard language and dialect: division of function, not replacement

Language planning offers a more useful frame than replacement: division of function. A standard language can carry cross-regional public communication. A dialect can carry local intimacy and community memory. It can transmit culture. This is not a division between a higher and a lower language. It is a division between domains. A healthy language ecology does not require killing dialects to keep the standard efficient. It requires each system to do its work.

That division needs institutions, not slogans. Public services can use a common language while remaining accessible to speakers who need support. Local languages can have room in family life, community events, media, and education. Schools can teach code-switching as a competence rather than treating dialect forms as errors to be corrected. Media can present dialects as complete registers, not only as comedy, rural color, or nostalgia. The key terms are access, domain allocation, and language rights.

Code-switching is normal in multilingual and multidialectal societies. A speaker may use the standard in a lecture and the dialect at dinner. That is not confusion or disloyalty. It is an ordinary skill. The either-or frame (standard or dialect) forces a choice that most speakers do not need to make. The functional frame lets both systems remain available.

6. Managing boundaries, not erasing them

The claim that dialects raise communication costs describes real friction. Its explanation and its remedy are wrong. The cost is a boundary cost, not a dialect cost. Ambiguity is a property of language, not a dialect defect. Standardization can lower some cross-dialect costs, and it can create new ones: language insecurity, the loss of local knowledge, and a narrowing of linguistic diversity that is hard to reverse.

A better position is not "standardize to avoid ambiguity." It is "build a functional division between standard language and dialects, and recognize the completeness and efficiency of each system in its domain." Communication does not aim at the elimination of all ambiguity. It aims at enough mutual understanding despite ambiguity. A standard language cannot achieve that alone. Dialects cannot achieve it alone. A division of labor can do better than either.

The practical task is to manage boundaries without pretending they can be erased. A society can ask strangers to use a common language in a hospital queue. It can also let a grandmother tell a story in the language she dreams in. The task is to keep both sentences possible.