It is in the nature of a rhetorical text that the author uses linguistic ’tropes’ (metaphors, similes, analogies, imagery, etc) to persuade a reader about the reasonableness of a specific standpoint. It does not build arguments that systematically lead to a stand point but instead tries to convince the readers (or listeners) by using the abilities built into natural languages. Such texts do not easily lend themselves to a discussion regarding the steps taken in developing a stand point, unlike, say, a mathematical proof or a philosophical argument. Of course, I do not find such texts ‘irrational’.
One possibility is that you do not expect to encounter such emotions in a rational discussion between two intellectuals. While you would be right on this score up to a point, please reflect on the possibility that you are circumscribing ‘rationality’ very narrowly: in such a circumscription, ‘rhetorics’ would not be considered ‘rational’ any more than discussions involving a combination of what we Indians call ‘rasa’. Taunts and teasing, humor and irony, sarcasm and ridicule, are as much parts of a rhetorical text as are praise and laudations. That is why I did not make the remark that Hogan’s text was a laudatio but used that character of his text to enter into a ‘discussion’, which, I admit, failed.
The idea that scientific theories should not use rhetorical strategies is something that no intellectual who knows anything about the relevant domains accepts. In a time when cognitive science, historians of science are studying the role that analogies, metaphors, and other linguistic tropes play in the process of scientific discovery (leave alone the propaganda of science), this idea belongs to yesteryears.
Note: When I say ‘rhetorical’, I do not mean anything pejorative; ‘Rhetorics’ was the general name for the art of persuasion and argument for nearly 2000 years in the West. Even to this day, in high schools in some European countries, a subject called ‘Rhetorica’ is taught to young people.