Word studies are an important part of the study of the biblical text. There importance can be over emphasized however. From time to time we encounter an "exegesis" that is basically nothing but word studies (or the author claims they are). But as valuable as a word study can be they can also be easily distorted and abused.
The abuse of word studies has actually come under some rather intense scholarly criticism over the last generation or so (led by James Barr). Some of the worst violations of sound rules for word studies are by scholars themselves . . . especially in the Theological Dictionary of the New Testament edited by Kittel. Most scholars today will say "TDNT is still a valuable tool but it must be used with caution." The same is true of Colin Brown's Dictionary of New Testament Theology. But the principles of a good word study are also violated routinely by preachers. What follows are seven guidelines for a procedure in doing a word study.
1) A distinction should be made between glosses (that is translation substitutes) and definitions. All of the meanings (please note the plural) of a particular English gloss may NOT transfer back to its Greek, Hebrew or Aramaic counterpart.
2) The study of a word is not equal to the study of a concept. This is, in the opinion of James Barr (i.e. The Semantics of Biblical Language) the fundamental problem with TDNT and many other biblical studies. A study of the article in TDNT on agape for example provides more than adequate criticism here. It is not infrequent to hear that "agape" means a distinctive kind of godly or Christian love. The exchange between Jesus and Peter in John is a classic case where this rule is abused (greatly!). This conclusion can only be reached by a highly selective reading of John's usage of the terms agape and phileo. Certain approaches to worship share in the problems of this "word study" approach to biblical theology. I would suggest that the ever popular "Edification Model" to worship exhibits these problems.
3) A single word often has several distinct . . . even UNRELATED senses. There is, often, no common core sense underlying all of its occurrences. (Please recall my illustration of the word "means" itself in the first installment of Text & Context.
4) The lexical sense(s) of a word should be studied using both the conventional conceptual approach which identifies the concept(s) the word denotes and the lexical field approach which compares and contrasts the word with other words with related senses.
5) Etymology is commonly useful as a mnemonic device for remembering transparent meanings, but it is generally irrelevant in determining a word's sense in a particular context. The word "automobile" means "self-moving vehicle" but few people actually know that or use the word because of that etymology. One example that is quite common is the word "ekklesia" to which preachers commonly will define by its "etymology" to mean "the called out." But there is little to no evidence that Paul or Luke ever actually used the word in that sense. Context, not etymology, determines the meaning of a word.
6) The immediate context is the most determinative feature for identifying the meaning of a word in particular occurrences.
7) The immediate context of word usually points to only one of its possible senses. Deliberate double entendre is rare and should normally be invoked only when clearly indicated.
The new third edition of Baur's Lexicon (BDAG) is an essential tool and should be consulted long before TDNT. The recent New International Dictionary of Old Testament Theology & Exegesis 5 volumes, ed. by W. A. VanGemeren moves ahead of the older theological dictionaries in avoiding many fallacies in word studies. For further reading on word studies I recommend the easily accessible material by Donald Carson in Exegetical Fallacies (a book that belongs in every preacher's library). If you are hardy then Peter Cotterell and Max Turner's textbook Linguistics & Biblical Interpretation is a must have.