Ordinary language is all right.
One could divide humanity into two classes:
those who master a metaphor, and those who hold by a formula.
Those with a bent for both are too few, they do not comprise a class.
'The correct method in philosophy would really be the following: to say nothing except what can be said...'
'I once wrote, the only correct method in philosophy consists in not saying anything...'
These dicta continue, of course. The interest of later philosophers is almost always in their continuations, in the ways they specify what you can say or how to critique what is said wrongly or the attitude to strike toward someone whose lot or good fortune it is to receive your critique. Authors who indulge this interest often seem to have an apologetic air, aware that the pages they fill trying to explain, or get clear on, the specifics are embarrassed by something else you can hear - something Wittgenstein brought out plainly in his paraphrase of himself - in that line from the Tractatus: 'not saying anything', 'to say nothing'.
I'm drawn to the idea of saying only what needs to be said. I find it hard to read almost anything and think that it needed to be said. It's not that I think there's nothing to say. I just rarely feel that I've read words that had to be said, that were called for, that someone was drawn to say, had to get out, needed to put on the page.
One of the most remarkable ways to write is to write to a friend. It's as if the friend frees your thoughts. Where you might otherwise have nothing to say, the expectation that somehow, your words would be reflected, or inflected, back toward you so as to surprise you, can make you eager, voluble. The friend makes your own words fresh, thoughts new.
The writing the world expects one to do is rarely directed toward a friend.
'And I want to say: The difference between real and imaginary, between existence and absence, is not a criterial difference, not one of recognition.'
Imagine someone saying: 'I can tell that this is a real thing, it's like the other real things I've seen before.'
Will Oldham's archaisms often add a touch of artifice to his songs in just a way that the narrators seem more real, present, because of their self-consciousness. Characters become singers playing characters. But that phrase in 'Darkness' - 'its dreadful and position' - lingers on like a thing.
Cavell interviewed in Bookforum, on 'underwriting':
'JR: Yes, in The Senses of Walden  and elsewhere you speak of Emerson and Thoreau as "underwriting" what has come to be known as ordinary-language philosophy. Can you explain?
SC: Well, first of all because they shun philosophical language. That is, they don't shun philosophy, but whatever they can't put in their own voice they don't. And as far as I was concerned, they were putting just about everything in their own voices, and I found that inspiring, unfashionable and inspiring, and partly one because of the other.'
'One of the few books I've ever had that was stolen - not by me, as it happened, but by a girl I persuaded to steal it for me - was William Carlos Williams' The Wedge. It proved fire of a very real order.'
Philosophers—perhaps only insofar as they are scholars, or worse, 'professionals'—like to talk about 'the arguments'. They ask, what are the arguments? And they size others up by seeing whether or not they 'know the arguments'.
Sometimes, in this, philosophers fancy themselves to be like mathematicians. But mathematicians ask for a proof, not the proof. They recognize that there may be other ways to establish a conclusion, and often would like to know them, not just for reasons of economy of assumption, but because the other proofs are nicer; more elegant; give more insight.
Philosophers relate to their arguments differently than mathematicians relate to their proofs. And to arguments not yet heard, as against proofs not yet found.
'Now was his last chance to see her; his plane left tomorrow.'
The individual episodes in a crime drama often—almost formulaically so—coincide with the stories of individual crimes, cases, investigations. The story of an investigation ends when the investigation does, for the most part, so the tendency is toward closed cases. Detective-focused narratives, whose stories of investigations tend to conclude with the identification of the criminal, say with his confession, seem particularly liable to portray criminal investigation as a kind of inexorable vindication of morality or worldly or eternal justice, so far as they only tell stories that end, cases that are closed by episode's end. Truth will out, as long as there is someone who will look for it, investigate. Beyond the constant presence of pragmatic obstacles to solving a crime, obstacles to the discovery of the truth show up as social, legal, political, economic, institutional—but these are usually only occasional or typical in a highly episodic narrative, however much the way that they characteristically appear in a given narrative can help determine its overall mood. The backed-up forensics lab, the proximity to powerful countervailing forces (government, wealth, public opinion, gangs, mutually mistrustful communities of people and their political representation), a climate of racism, complacency, Hobbesian economic conditions, corruption more or less endemic. The more episodic a crime drama is, the more the most proximate obstacle to discovery of the truth—or even the search for it—is likely to be personal, in the person of the detective himself; in his struggle against a personal failing, or in the conflict his person or personality generates between him and those he must convince with his investigations (or, convince to let him investigate at all).
The more space a detective-focused episodic narrative makes for the participation of the legal system, the more potential there is to unsettle that narrative's tendency toward the vindication of righteousness. The detective finds out who did it, but then it's out of his hands, the case turned over to the justice system. Or the detective knows who did it, but can't prove it in court (or even has no hope of ever being able to). Or isn't given an opportunity to see the case forward (for political reasons, because people would rather let sleeping dogs lie, because even the victims are more comfortable not pursuing justice). Or the detective must himself take part in the post-investigation proceedings, and compromises his own case through his personal failings, his history, when they are made the issue. If the legal system (or some other system) plays a genuinely independent role in the narrative, more than occasionally thwarting justice, it's likely to seem adversarial to everyone in the narrative assigned to investigating the truth. Much seems to depend on the extent to which the legal system enters into the same one-case-per-episode structure as the criminal investigation does.
Episodic crime drama often foregoes supra-episodic narrative, opting at most for what I would rather call intra-episodic narrative. Usually thematic, too; personal, romantic, family, spiritual life. Or incidentals of the job which can easily be emphasized to the point of making the crime drama a workplace drama. Strangely, a narrative of professional advancement is well-suited for supra-episodic use but seems underexploited and underdeveloped. Some examples to consider: Andy Sipowicz, Vic Mackey, Tim Bayliss, Cagney & Lacey, Lester Freamon, Mike Logan, Mulder & Scully. Professional setbacks are more common (and cheaper, in the economy of the overall narrative). The rookie or quasi-rookie role often emphasizes this dimension but seems not to invite centrality or prominence (more likely one spot in an ensemble, as on Homicide or Southland), as if whatever a 'new' character in an episodic drama is new to, it can't be the investigation of crime. Dean Winters' SVU character doesn't last long. The characters of Scully, Vera Farmiga's Susan Branca on Touching Evil, Rachel Nichols's ingenue on The Inside, are experienced investigators but their newness to their assignments is exploited for dramatic, thematic, psychological purposes.
Supra-episodic or intra-episodic narrative centered on a relationship is common, probably because it's so adaptable, and the structure of an episodic crime drama leaves so much about the character of the investigator, and his or her relationships, free to be determined, unconstrained by the logic of the crime investigation narrative. For some reason romantic relationships between partners or co-workers of the same job and status in it seem uncommon, used only transiently; this might be because most shows do not have the narrative resources to tell non-stereotyped or nongeneric relationship stories, only stories that are thematically and structurally subsidiary (justice comes first, love will get you killed, this job asks too much of you, we have to keep this a secret from the others, or the boss). NYPD Blue might be a good exception here, and it's interesting that it codes as 'serious' or 'adult', perhaps just because its characters are allowed to have relationships with those they work with to satisfy personal needs under the constraints under which people in certain lines of work often form relationships, and those relationships aren't (necessarily, though some are, contingently, like the Franz-Lawrence relationship) overdetermined by the structural or thematic characteristics of their jobs or narrative roles (cop versus lawyer, professional and detached versus warm and humane, instinctive versus methodical). Romantic relationships in other shows are more typically contrastive, and highly underdeveloped unless both members are crucial to the repeating, episodic component of the narrative. To complement the regularity of closure characteristic of highly episodic narratives, a supra-episodic will-they-won't-they narrative is common, but can be well developed depending on how closely integrated with the repeating investigative narratives it is. The relationship narrative on a show like Bones or Castle is cheap, only extrinsically related to the unfolding of a given episode's investigation. (Where Castle can obviously only last so long once Beckett and Castle go for it, since the narrative lacks the resources for them to join, part, and continue to work together, the dogged continuation in Bones of the story after its will-they becomes a they-did makes it seem indulgent, comfortable, an exercise in fantasy fulfillment.) The relationship on Standoff between partnered hostage negotiators with contrary styles and professional competencies permitted a supra-episodic narrative that had an intrinsic connection to the drama of each episode.
The open investigation or unsolved case is a special counterpart to the closed one in an episodic crime drama, and it often serves a supra-episodic narrative role, especially when individual episodes correspond more or less to individual (and thus routinely solved, closed) cases. An investigator often has a personal connection to such a case, but (for some reason) it's likely to be a legacy rather than originally the detective's own case: investigation of the death of family or friends. Since the case has to pre-exist the narrative, or come into existence with it, it is characteristically set in the more substantial past, and the detective's involvement in it relates them to their own past, or to characters who know the past but for whatever reason won't reveal it to the detective. Often the detective can double as a victim in this capacity. (Charlie Crews in Life, Veronica's rape and the disappearance of her mother on Veronica Mars.) The past case is sometimes used to define the character biographically ('this is what made me become a cop'), but more often psychologically (Monk's Trudy, the abduction of Mulder's sister Samantha, the victimization of Rebecca Locke by a serial killer from whom she escaped—the last being a closed case but one which remains 'psychologically' unresolved). The detective himself can be the subject of the open case, like Vic Mackey on The Shield. There, there is an open case from episode to episode because he is the criminal and hasn't yet been exposed by others. His own work as an investigator in each episode correspondingly seems more like the work of an operator (who makes things happen) than of a detective (who finds out what happened).
Where the routine episodic closure of cases is dropped, an open investigation or unsolved case can define the entire narrative: Twin Peaks, The Wire.
An open investigation serves especially well as a supra-episodic narrative frame in combination with a supra-episodic relationship narrative where the relationship is between a detective and a criminal. Will-they-or-won't-they is cast in terms of loyalty and betrayal (against the demands of justice and friendship or love) rather than in terms of romantic union.
One thing that clearly stands out in David Simon's book in contrast to the dramatized version of it is that the detectives are given biographies. Their stories, backstories, are told, roughly as they come into prominence because a case they are working has also been described. These biographies overlap with the backstories of the TV characters, but whereas the biographies in the book are interpolated, in relatively whole form (as far as they go), into Simon's account, the characters on the show have to tell, or reveal, their own life stories—and they do so only incidentally and intermittently, to their fellow detectives (or occasionally to witnesses, suspects, romantic prospects, strangers, less often families), in connection with a case or in the conversations that fill up the spare moments that come while working one. In the book, Pellegrini is given a long history of dissatisfaction with his work, with his direction in life. On the show, Bayliss is given some of the same story, like the time spent on the mayor's security detail specifically in order to secure an assignment to the homicide unit. This comes out naturally, given that it's a salient biographical fact for any detective new to the force; it's what he tells the others when they ask where he was before. Other facts of Pellegrini's biography that do or do not, to some degree, show up elsewhere in Bayliss's backstory don't come out then, though, because relation of Bayliss's biography is controlled by the dramatic situation, by its narrative form, rather than being controlled by a journalistic narrative that permits arbitrary suspension, retrogression. (A different show could have used a different form of narrative.)