Okay before I continue this oh so fascinating series (all joking aside, if you only read one post in the series, this is the one to read), I should clarify two things.
(1) The semiotic approach being described here is particular to Umberto Eco. This series should actually be called “Eco’s Semiotics for Dummies” because I am not providing a general survey of multiple views (although some of those views are dealt with [or presupposed] by Eco). As well, continuing the “for dummies” theme, I will mention that http://www.thefreedictionary.com was a great help to me while reading through this book. See a word you don’t know? Well, chances are, I didn’t know it either.
(2) One of the things I struggled with the first time I worked through this book (I’ve gone through it twice now) was how the chapter tied together. I had some trouble understanding the flow of his argument from section to section. How does it all tie together? So, let me clarify a bit of where we’ve come from before I move to the next section. In the introduction (Part I of this series), Eco argued that he wanted to resolve the dichotomy between the sign-function and semiotics by arguing that the semiotic process applies to signs themselves. In order to do to this he must locate himself within a range of approaches to hermeneutics. Thus, he posits a continuum with x at one end and y at the other. At point ‘x’ are those who only see one meaning in a text, and at point ‘y’ are those who see any meaning in a text. Eco’s is interesting locating himself between those two points and positing a limited range of meanings in a text. Therefore, in Chapter One (Part II of this series), Eco further explicates the positions of those at the points ‘x’ and ‘y’ when dealing with the basic semiotic notion of the sign. He then explains his middle way in relation to this concept, thereby affirming his thesis. Therefore, having affirmed this via media, he goes on to explain just exactly how he understands meaning. And this is the topic of the section we will now look at. Thus without further ado:
 Dictionary vs. Encyclopedia
Eco’s central thesis in this section is that the Porphyrian tree (i.e. dictionary) model of definition is untenable. Rather meaning must be understood through an encyclopedic model, and the Porphyrian model should just be employed as a pragmatic tool. Okay, far enough, but what does this all mean?
Well, we need to first understand a dictionary approach to meaning. Hjelmsev was probably one of the first to propose the idea of the modern dictionary by extending his sign/figurae distinction (see Part II) from the expression-plane to the content-plane as well. However, this approach requires a limited number of primitives, and so a lexical system of hyonyms and hyperonyms is created in the format of a tree that slowly tapers at the top (i.e. Regnum, Class, Subclass, Order, suborder, Family, Genus, Species, Common Name). Such an approach to definition does not drift into synonymy but is using short-hand expressions to signify common sets of properties. Properties are contained or meant by certain taxonomic labels. However, for this approach to be successful there must be a limited number of primitives (“primitives being the things at the bottom of the tree) and a criteria that firmly establishes what is a primitive and what is not. Unfortunately, this approach has trouble resolving those problems.
Before pressing onto to his proposed resolution, Eco first takes a step back and looks at where this dictionary model came from. It was Aristotle who first argued that definition meant establishing a set of essential attributes. He also developed the notion of predicables, the modes or categories which can be applied to a subject. Porphry developed Aristotle’s notion and posited five predicables (genus, species, differentia, property, accident) and this way of thinking has impacted definition ever since.
As mentioned above, this approach is problematic for it assumes a finite set of genera and species which are mutually definable through hyonymy and hyperonymy. The reason why this is so problematic is that such a system of clusters is completely arbitrary. It is only arbitrary differentiae that make up the various clusters, and so the tree can be continually re-elaborated and rearranged. Furthermore, because differentiae are accidents, they are also infinite, or at least indefinite. Therefore, the Porphyrian model of definition is untenable. The tree dissolves.
However, the notion of defining things by clusters of common properties is useful to the extent that it reveals a dictionary’s true identity: a dictionary is a disguised encyclopedia.
An encyclopedia contains the world knowledge of a given culture. This encyclopedic knowledge provides frames and scripts for any given situation, that allows a certain meaning to be the appropriate meaning for that context. Within an encyclopedia there is no hierarchy of properties, but such hierarchies are only created ad hoc in any given context. Therefore, the encyclopedia is the regulative hypothesis that allows local speakers to construct an arbitrary dictionary in order to ensure communication. Encyclopedic knowledge can be mapped as a net-like labyrinth where all points can be connected with all other points to an infinite degree. Thus, it is a myopic algorithm for one can only see where one is at a specific local point, one cannot see the whole at any time. Thus, “truth” cannot be discovered in any sort of global way, it can only be recognised in local places — with the recognition that that “truth” always has an ideological bias.
Therefore, Eco concludes, the encyclopedia is a semantic concept and the dictionary is a pragmatic tool. How exactly the relationship between an encyclopedia and an ad hoc dictionary plays out will be made clear in the next chapter.