Peircian Semiotics

Charles Sanders Peirce (1839-1914), the American philosopher, logician, and scientist (among other things), coined the term "semiotics ", and introduced several of its fundamental concepts. Peirce (pronounced "purse"—the family name was earlier spelled "Pers") declared that anything can be a sign—words, images, odors, objects, ... anything—so long as someone, in some situation, interprets it as standing for something other than itself. In other words, by themselves, these tokens have no meaning; they are given meaning (and thereby become signs) only when they are interpreted as such. Of Peirce's ideas, the most relevant to UID are his model of the sign (the semiotic triad), and his system of classification of signs (based on how they convey meaning).

The semiotic triad

The parts of Peirce's triadic model of the sign are:

• a representamen, the (not necessarily physical) form of the sign;
• an interpretant, the sense made of the sign in the mind of the observer (this can be another sign); and
• an object, that to which the sign refers.

These three parts are together known as the "semiotic triad", and together they form a sign. As mentioned before, meaning is not directly attached to the sign; instead, it is mediated through the interaction between the representamen, interpretant, and object. Peirce referred to this as the process of "semiosis". The figure to the left displays these relationships (the dashed line represents the indirect path between the representamen and object).

As an example, consider the notorious Microsoft Windows "Blue Screen of Death" (the representamen). When a user encounters this error code screen, she may think that her computer has crashed (the interpretant), and press Ctrl-Alt-Del to reboot the machine (the object).

Especially important in this model of the sign is the presence of the interpretant as an explicit component. Since the interpretant is created by the observer, the object is not given, but inferred. This makes a sign's meaning highly dependent on context. With two different interpretants, a single representamen (or token) can have two different meanings. Similarly, a single object can be represented by two different tokens. In both cases, we are dealing with two different signs.

The figure below shows an example of a representamen with two different interpretants. Consider a user faced with the common "Press any key to continue ..." prompt (the representamen). In one case (i.e., one sign), this causes the user to think of the Enter key (the interpretant), and results in her pressing that key (the object). In the other case, the same prompt causes the user (a computer novice) to think of the nonexistent "Any" key (the interpretant). As can be expected, this results in his futilely searching for the key (the object), and eventually leads to confusion and/or frustration. (Though the particular situation illustrated in this example is unlikely to occur, it still serves to highlight the necessity of knowing your users when designing an interface.)

Peirce's classifications of signs

Based upon how they convey meaning (i.e., the relationship between the representamen and the object), Peirce classified signs as icons, indices, and symbols (note that like "sign", these terms do not refer to their common usage).

• An icon is a sign in which the representamen is perceived as resembling (i.e., having some of the qualities of) the object. In Peirce's words, such a sign "refers to the Object that it denotes merely by virtue of characters of its own ... such as a lead-pencil streak as representing a geometrical line."

  For example, the trashcan "icon" (in computer-speak) is iconic of an actual trashcan, and the action of moving a file "icon" into the trashcan "icon" is iconic of the actual action of putting something in the trash. Also, the sound effects (usually) played when buddies sign on and off of a chat client are iconic of the actual sounds of a door opening and closing respectively.

  A diagrammatic icon is a type of icon that preserves the geometric properties of the object; in this sense, the resemblance is not necessarily based on their similarity in appearance, but on the relationships between their parts. For example, the trashcan "icon" discussed previously would be better described as being diagrammatic iconic.

• An index is a sign in which the representamen is in some way (i.e., physically or causally) directly connected to the object. Unlike an icon, an index has no significant resemblance to the object. Instead, there exists a law-like relationship between the representamen x and the object y: if x, there always or usually is a y (it is for this reason that unedited photographs and video—which are indexical of the effect of light on the film—are often admissible as evidence [1]).

  For example, the earlier mentioned trashcan "icon" is indexical of the process of deletion; similarly, a diskette "icon" is indexical of the process of saving. The "Blue Screen of Death" is indexical of the computer having crashed. Also, a person's username and password are together indexical of that person (and hence can be used for authentication).

• A symbol is a sign in which the relationship between the representamen and the object is arbitrary or conventional (i.e., it must be learnt). According to Peirce, such a sign "is constituted a sign merely or mainly by the fact that it is used and understood as such."

  For example, language is generally considered to be symbolic. Logos (such as AOL's Running Man) are symbolic of the corresponding company. Also, using colors (red, yellow, and green) to represent the quality of connection (poor, fair, and excellent respectively) in a network status "icon" is symbolic.