Introduction

Finding Web Metaphors
A Proposed Corpus Search for Metaphors of the WWW
by Zach Tomaszewski

Summary: A study to search both a large general-use corpus and then small Web-specific corpora for example sentences (citations) containing Web metaphors, as evidenced by keywords from the target domain. In this way, document the existence of any unique Web metaphors found, as well as sample source domain keywords.

Literature Review

Lakoff and Johnson's Metaphors We Live By (1980).

• not simply figure of speech
• based on language, conceptual system is metaphorical
• partial mapping between domains; thus, multiple source domains/target (ex: ideas -> people/life, plants/growth, commodities/products/resource, fashion; time -> resource, event-structure)
• Also some feature hiding/downplaying
• result of our embodied experience (physical experience more clearly delineated)
• coherent system of metaphors shapes how we act and what we consider true

Literature Review (continued)

• We convey system models through interface metaphors
• The COMPUTER IS A DESKTOP metaphor: files, hierarchical organization, trash can
• shapes our interactions and expectations
• poor mappings: removing a Mac floppy
• feature hiding: documents verses programs, contiguous files, permanent deletion
• yet metaphor often not dropped later (as vehicle)
• So metaphor vital to understanding and communicating about abstract systems

Research Question

Yet, no extensive study on what conventionalized metaphors exist for the Web.

Studies on utility of certain metaphors, or discussion of ramifications of certain metaphors, but no broad scope. Personally constantly afraid we're overlooking something.

Best way to do this: archival study (corpus research).

Research Design

• General corpus (ie, Bank of English)
• Identify keywords of the target domain, the Web.
  (trickier than expected.)
• Search for keywords, returning sentences (citations).
  (randomly sample if necessary)
• Analyze sentences for use: literal, conv. metaphorical, novel metaphorical or homonymous.
• (Traditionally in linguistics, then determine frequency of keyword as metaphor in language, or conversational context. Skipping that part. Would like relative frequency, but shaky ground.)
• Repeat with specific corpus (websites?).

Expected Contributions

• Open the way for research on specific metaphors:
  --Web use
  --cyberspace/frontier legal issues
  --cross-cultural comparisons
  
• Web corpus generation -> change of metaphor over time and intended audience context. (But only measures author use.)

Questions?