Primary and Secondary Indicators                        John Black


                Informal logic texts differ in the treatment of certain 
        words and phrases used by arguers and argument analysts alike to 
        locate premises and conclusions in arguments.  Most agree on 
        "therefore" and "because" and their synonyms; the disagreement 
        concerns such words and phrases as "firstly," "in the second 
        place," "on the one hand...on the other hand," "furthermore," 
        "moreover" and, in fact, "in fact." 
                An exhaustive search of the literature on my bookshelves 
        reveals some of the diversity.  Johnson & Blair's Logical Self-
        Defense, Govier's A Practical Study of Argument, Fogelin's 
        Understanding Arguments and Kelley's The Art of Reasoning make no 
        mention at all of these words and phrases in their lists of 
        indicators.  Thomas' Practical Reasoning in Natural Language and 
        Chaffee's Thinking Critically include those involving ordinal 
        numerals, while Freeman's Thinking Logically treats "moreover" as 
        a premise-indicator in an example on p.205.  Wilson's The Anatomy 
        of Argument is by far the most inclusive; it mentions three 
        classes of locators (equivalent to indicators) other than the 
        synonyms of "therefore" and "because." 
                Wilson's three classes are (a) the numerical devices 
        mentioned above, (b) devices used to indicate an accumulation of 
        considerations, including the non-numerical ones mentioned above, 
        and (c) contrastive devices such as "but" and "however."  He calls 
        them all "specific premise locators," contrasting them with the 
        synonyms of "because," which are called "general area premise 
        locators."  While Wilson is right in mentioning these devices, and 
        in asserting that they play a role in the analysis of arguments, I 
        think his treatment of them is incomplete. 
                Firstly, the same devices may be used to enumerate 
        conclusions: they are not essentially premise-indicators.  
        Secondly, students can easily become confused about how much they 
        indicate, taking them as a sure sign that a passage contains an 
        argument when in fact it doesn't.  Though this is a feature they 
        share with such equivocal indicators as "because" and "since", the 
        non-argumentative contexts in which the latter occur are more 
        clearly assigned an alternative interpretation.  I want to suggest 
        a way to complete Wilson's treatment in a way which brings greater 
        clarity to argument identification and analysis. 
                Let us give the title "primary indicators" to the synonyms 
        of "therefore" and "because", and the title "secondary indicators" 
        to words and phrases in the other group.  Primary indicators have 
        an absolute direction or sense: they are conclusion- or premise-
        indicators, according to which role they indicate for the 
        statement following them.  Secondary indicators do not have an 
        absolute sense, though they may have a sense relative to some 
        features of the context in which they occur.  In most cases, 
        secondary indicators function as indicators of argument structure 
        only to the extent that they occur in a passage in conjunction 
        with some primary indicator.  The sense of the primary indicator 
        and the relative position of the secondary indicator determines 
        the sense of the latter.  
                There are exceptions to this rule (my fourth paragraph 
        contains two!) when there is no relevant primary indicator.  In 
        these cases, however, the interpretation of the passage as 
        argumentative and the ensuing analysis of its structure must rest 
        on features other than the presence of the secondary indicators, 
        and the determination of the sense of the latter must be made in 
        relation to a prior construal of the overall structure of the 
        argument.  
                Hence, on the one hand Wilson's account has gone beyond 
        those of most authors in drawing our attention to some important 
        devices, but on the other hand it is not general enough in its 
        application.  Furthermore, this very paragraph contains three 
        examples of the limitations of the account.