If one wishes to attain a goal then it is necessary to first ascertain what that goal is. The foregoing is as obvious as it is frequently overlooked. In his chapter 'Sermon Conclusions' W.E. Sangster seeks to correct this oversight; and this one observation alone is worth the time taken to read the chapter. The remainder of the section contains five possible approaches to concluding a sermon and these approaches are the subject of the bulk of this paper.
Sangster's apology for the sermon conclusion is more poignant for a Christian message than it would be for a general address. Sangster, a leader of a church movement, was keenly aware that the weekly sermon needed to do something. If a church was to be alive, growing and healthy then the flock needed to be alive, growing and healthy and it needed to take at least some of its direction from the pulpit. Sangster's position was that many people seek to expound the scriptures but do not really understand the end they were attempting to achieve with any given sermon[1]. He argued that it would logically follow that they would not reach any particular goal. His retort for those that argue that the 'Spirit must lead' was that the Spirit is just as capable of leading during sermon preparation as it is during a message. Aligned with this thought Sangster pleads for a sermon to be sufficiently well prepared that adequate time is left for the conclusion; 'clock issues' he considers a symptom of inadequate preparation.
Having pled for the value of a conclusion Sangster proceeds in characteristic fashion to categorize the types of sermon conclusion, chief characteristics of them and the principle application of each.
Sangster concludes his section upon conclusions with some pitfalls to beware; most of them center round distaste for an obvious and disjointed conclusion. Thus he denigrates 'the moral of this story is ...' and even 'in concluding' or 'finally' especially if the latter is delivered more than once. He particularly stresses that if a preacher announces the time remaining in a message ('just a minute more') then the preacher must be true to his word or the congregation will take it as a signal of a 'run on' message. In similar vein Sangster cautions against adding new material to a conclusion that would detract from the focus of the rest of the message. Sangster goes as far as to advocate word-smithing and even memorization of the conclusion of a message so that it can be delivered with pace and precision. In slight counterpoint he cautions against endings which are banal (today we might say 'sound bites') noting that such an ending can seriously undermine the gravitas of the message.
For me personally Sangster's methodological analysis of the conclusion almost undermined the key point which is that we should know what we are trying to achieve. It would be a shame if an entirely valid end-goal got lost in the study of the route to its achievement. Notwithstanding recapitulation and application are both entirely valid methods of conclusion that have served throughout the ages. Recapitulation is most appropriate where the body of the message has been balanced and one wishes to 'tie the ends together' to aid memorization and integration for the audience. Application is most appropriate where the message has followed a route whereby the applicability is not yet clear. Demonstration, Illustration and peroration all presume that it is good to leave the congregation with something man-made rather than directly Scriptural; it is hard for me to think of a particular 'goal in mind' that would make that seem like a good idea.