A statement is ambiguous when it can be given two or more unrelated interpretations.
Ambiguity in natural language
Ambiguity extends beyond the domain of language, but its most well-known varieties are linguistic:
- Lexical: due to a word having multiple meanings (polysemy)
- Syntactic: due to the arrangement of words having multiple meanings
- Semantic: due to an expression having multiple meanings
- Endophoric: due to a reference that could refer to multiple parts of a text
Some of these terms are closely related; for example, lexical ambiguity is a type of semantic ambiguity and Semantic ambiguity may be an effect of syntactic ambiguity.
Ambiguity outside natural language
Ambiguity is not restricted to natural language.
A painting, button or equation can be ambiguous.
$$\frac{si\cancel{n} \ x}{\cancel{n}} = six = 6$$
In general, an ambiguity involves a one-to-many relationship between something that is perceived and the set of its possible interpretations.
Disambiguation is the process of reducing this one-to-many relationship to a one-to-one relationship, between the what is perceived and its intended and/or correct interpretation.
Productive ambiguity
In many cases, ambiguity is bad.
An ambiguous instruction can be catastrophic when operating machinery, performing a surgery, or preparing a meal from a potentially toxic fish.
Ambiguity, however, can also be useful.
When a recipe prompts us to add salt “to taste”, it is ambiguous with respect to the quantity of salt. Yet it also provides us with the agency to choose and experiment with the quantity.
Movies with enigmatic endings can be frustrating but they also arouse our curiosity and provoke discussion.
Ambiguity and discovery
Even in a methodical discipline like science, ambiguity can be understood as a necessary stage of that activity.
Any scientific discovery begins with an ambiguity — a phenomenon that admits of multiple possible interpretations. This prompts the scientist to seek the interpretation that has the highest probability of being correct.
Great scientists seek out ambiguities with the same vigour as they do answers.
Technological ambiguity
Technology can be ambiguous in different ways:
- Functional: multiple interpretations of what it does
- Procedural: multiple interpretations of how to use it
- Mechanistic: multiple interpretations of how it works
A technology may have multiple valid functions and different valid procedures for its use. This may result in the perception of ambiguity for the user, even if the design is intentional and useful. As one becomes skilled, the sense of ambiguity may be replaced with a sense of possibility.
Natural ambiguity
Naturally-occurring things or events may be ambiguous.
While enzymes were classically considered to be substrate-specific, it is now known that some enzymes exhibit substrate ambiguity.
The same phenomenon has also been observed for molecular chaperones.
Semiotic ambiguity
Natural signs may be ambiguous. A classic example is smoke in the distance, which can be interpreted as an indicator of fire but also the presence of humans.
The smoke may have been caused by lightning igniting a dry log or by the campfire of a hostile group who is coming to get you.
Ambiguity and generalisation
Both ambiguity and generalisation involve one-to-many relationships.
- Ambiguity: many possible interpretations
- Generalisation: many possible instances
The distinction is that the interpretations in a case of ambiguity ($A$) are mutually independent, whereas the instances in a generalisation ($G$) are members of some class.
$$A: a \rightarrow \{x, y,…, z\}$$
$$G: g \rightarrow C, \ C = \{g_1, g_2,…, g_n\}$$
A generalisation assumes some model that explains the class and its membership.
Without an understanding of the model, a generalisation may have the appearance of an ambiguity.
Quotes
Tell all the truth but tell it slant
— Emily Dickinson