A tangled web

« previous post | next post »

Daniel Deutsch sent in this quotation

“The Attorney General has previously stated that the Special Counsel repeatedly affirmed that he was not saying that, but for the OLC opinion, he would have found the President obstructed justice. The Special Counsel’s report and his statement today made clear that the office concluded it would not reach a determination — one way or the other — about whether the President committed a crime. There is no conflict between these statements," a joint statement from DOJ spokeswoman Kerri Kupec and Mueller spokesman Peter Carr said.

with this comment

I understand that Language Log is not a political site, but this calls for a language expert:

Affirmed that—was not saying—that, but for—would have found. No conflict.

I had to read it 20 times to understand it.

Indeed. The first sentence alone confronts us with a stack of five recursively embedded sentences, which would have been unusual even in the 18th century:

Definitely a fitting contender for the Trent Reznor prize.

And the preposed adverbial in that sentence's final clause — "but for the OLC opinion" — adds that extra je ne sais quoi: an archaic construction that's crucial to the meaning of the whole statement.

The second sentence only has four levels of Russian-doll embedding:

But four levels is still really unusual in contemporary writing. And the conjoined complex subject of the top-level sentence, the negative irrealis "would not reach a determination" two levels down, the interpolated adverbial "one way or the other", and the bottom-level whether clause, all combine to make this second sentence another challenging feat of  comprehension for the reader.

And then we get the simple but devastating envoi: "There is no conflict between these statements."

All in all, a masterpiece.

Update — I should clarify that my correspondent was this Daniel Deutsch, not this one, or this one, or any of the many other people by that name.



  1. J.W. Brewer said,

    May 30, 2019 @ 8:25 am

    I don't disagree with the overall assessment of unusual syntactic complexity, but that particular "but for" construction is, in this context, not at all an archaicism, although it may well be jargony. I.e., it's alive and well in the lawyer-jargon variety of AmEng. Indeed, the google n-gram viewer shows a significant increase in the later 20th century in the derived NP "but-for causation," which I imagine is even more opaque to the outsider. It's possible the same construction was previously widespread in non-legal use but subsequently became archaic there, but it also appears that a lot of hits for the string "but for the" in all time periods just aren't that construction,* so a more fine-grained analysis would be necessary to assess that.

    *Example 1 from 1883 (the construction in question): "This increase would have been considerably larger but for the drought which prevailed in 1881, continuing during the first half of the past year."

    Example 2 from c. 1930 (something else entirely): "For the white population births exceeded deaths in every State, but for the colored the births were outnumbered by the deaths in 10 States."

  2. J.W. Brewer said,

    May 30, 2019 @ 8:35 am

    I did a comparative search in the google n-gram viewer after hitting post on the prior comment and it may be relevant that the decline over the 20th century in the incidence of "but for" coincides with a notable increase in the incidence of "except for" which might outside of legal jargon be a pretty close substitute for the relevant sense of "but for." The third alternative I could think of for that sense is "if not for," which turns out to be consistently much rarer than the other two.

  3. Theophylact said,

    May 30, 2019 @ 8:58 am

    I think you meant "which would have been unusual even in the 18th century".

    [(myl) Thanks — fixed now.]

  4. Robert Coren said,

    May 30, 2019 @ 9:49 am

    What makes the first sentence especially challenging is the comma after the third that, offering the possibility that it's acting as a pronoun (as in, "I didn't say that!"), rather than introducing yet another clause. (The lack of an antecedent can be attributed to a presumed unprovided context.)

    (Pardon me if I'm not using the correct terminology; I'm the merest amateur linguist/.grammarian, but I hope my intent is clear.)

  5. Jonathan Smith said,

    May 30, 2019 @ 10:18 am

    It might be a tiny bit clearer without "he was not saying that", and it's not immediately totally obvious what shade of meaning this is adding… but it must be for reasons, e.g., perhaps the implication is "…the Special Counsel repeatedly affirmed orally to the AG (?) that in the written Special Counsel report (?) it was not meant to be implied that XYZ."

    But the underlying point is true that beyond not bringing charges due to particulars, Special Counsel clearly does/did not want even to comment or appear to take a position on whether certain behavior was even "charge-worthy" in some more general sense, which is exactly what 400+ (?) former DOJ prosecutors apparently felt really needed to be said. If one was "political", one would want a precise account of timing and basis wrt the emergence of these particular stances.

  6. GeorgeW said,

    May 30, 2019 @ 2:08 pm

    But for the law and facts, no collusion, no obstruction, right?

  7. ktschwarz said,

    May 30, 2019 @ 3:24 pm

    That quotation frame has a bad case of
    non-inversion: it would certainly read more smoothly as

    "… There is no conflict between these statements," said a joint statement from DOJ spokeswoman Kerri Kupec and Mueller spokesman Peter Carr.

    Perhaps the Washington Examiner aspires to the New Yorker's style.

  8. BobV said,

    May 30, 2019 @ 3:51 pm

    I loved the "Russian-doll embedding" reference, as that is the general topic at issue here.

  9. chris said,

    May 30, 2019 @ 7:02 pm

    But wait, there's more!

    There's no conflict between the *statements*, but the second statement undermines the potential rhetorical purpose of making the first statement (although I'm not sure if it technically qualifies as an implicature either).

    Mueller wasn't saying the but-for part because *he had already formed a policy that he would never under any circumstances say that*, so it is not justified to infer he believed it wouldn't be accurate. The existence of his no-comment policy makes it impossible to discern his true opinion, which is the whole point of the policy.

  10. JPL said,

    May 30, 2019 @ 9:42 pm

    It looks like the quotation is from DOJ spokeswoman Kerry Kupec; it would be helpful to also have the original quotation from Bill Barr's answer to a reporter's question, as well as a quotation from Robert Muller's statement or from the Muller Report.

    Bill Barr is a slippery character. My impression is that he always uses language very carefully: from his legal training he has the ability to make statements that are clear, precise and unambiguous, but he also often misuses his abilities to make statements that are intentionally ambiguous, vague or confusing, while maintaining plausible deniability. The key sentences in the infamous Barr letter were all ambiguous, intentionally so, where the more favoured interpretation was misleading, while leaving open a less obvious interpretation that more or less corresponded with the facts. From this fact one can conclude that Bill Barr is not operating in good faith.

    In the above quotation there are two relevant clauses:

    a.) "… the Special Counsel repeatedly affirmed that he was not saying that, but for the OLC opinion, he would have found that the President obstructed justice."
    b.) "… that the office concluded it would not reach a determination — one way or the other — about whether the President committed a crime."

    (a) is technically true, because, indeed, the Special Counsel was not saying that, but it is misleading. (b) is almost but not quite technically true, but it is also misleading. They are misleading because they lead us to believe the situation referred to is other than what it really was. From the Muller report and Muller's statement yesterday, the the fact of the matter appears to be that, after some consideration, from the outset, at least logically if not temporally, Muller decided that they would recognize the OLC opinion as valid and constitutionally binding on the form their report of results would take. In other words, they determined that their judgment, or refraining from judgment, about making a criminal indictment would be dependent upon the general principle of the OLC opinion. However, the OLC general principle did not prevent them from indicating a judgment that no crime had been committed, and Muller indicated that if that had been the case he would have said so. Muller also indicated that, according to the constitution (plus the OLC opinion), it is congress that has the role of arriving at the judgment of whether, given the evidence, a crime has been committed.

    (a) is misleading in the following way. The "but for" construction introduces a counterfactual condition, so that the following expression would be equivalent in core content to (a): "If the OLC opinion had not been there, Muller would have found the President obstructed justice (i.e., committed a crime)". This distorts the logic of Muller's actions: The OLC opinion was a fundamental assumption; therefore the question of "finding" a crime at the conclusion of the process did not arise for Muller. (b) is misleading in that the use of the word "concluded" could be taken to indicate a decision at the end of the process after all the evidence had been presented, and one not dependent on the OLC assumption. In fact this decision, or "conclusion" occurred logically, if not temporally, at the outset of the investigation, so that at the end no judgment of indictment was possible. However, a judgment of "innocent" was still possible under the OLC assumption, so clause (b) is false. Maybe one could say that there is no conflict between (a) and (b), but it doesn't really matter because they are both misleading, and it is a further indication that Bill Barr and the Republican party are not operating in good faith. It looks like none of Bill Barr's key statements should be taken at face value. The public should not let this man get away with his attempts to "spin" reality.

  11. JPL said,

    May 31, 2019 @ 2:16 am

    I should add that there would be no conflict between (a) and (b) (above) only under the misleading interpretations of the statements: i.e., only if the statements are understood as referring to a situation where Muller's decision not to indict or "find" criminal activity was not at all dependent on the assumption of the OLC opinion, but rather freely made at the end of the investigative process and dependent only on the indeterminacy of the evidence. But this is not at all the actual world.

  12. Andrew Usher said,

    May 31, 2019 @ 8:00 am

    Putting aside all politics, I find nothing wrong with the statement at all; it appears to be well-written and says exactly what they meant to say – which could not be said precisely with fewer layers. I understood it the first time (unless I'm wrong …) and did not find it tortured or obfuscated.

    The statement does not mention the time when 'the office concluded …' but indicated that it was in the past, so there really is no inconsistency. This omission doesn't seem to be any attempt at cover-up.

    k_over_hbarc at yahoo.com

  13. Ray said,

    June 1, 2019 @ 6:07 pm

    like andrew usher, I didn't have a problem understanding the statement, either. (I did have to look up OLC, which probably shows you how much I've been following the whole subject haha)

  14. jamesdb said,

    June 3, 2019 @ 6:38 pm

    I don't find the statement difficult to parse (perhaps due to a bad case of legalese). The issue this statement clears up concerns the following distinction.

    It is not true that the OLC Opinion forced:
    (a) the SC to not indict.
    (b) the SC to not to decide whether to indict.

    Before this statement, I understood/assumed Barr to have claimed (b), but now (and after re-reading his previous statements) it appears he is saying (a), which is not in conflict with Mueller's claim.

    it's striking to me, however, how strongly political sentiment affects interpretation. To those opposed to Trump et al., Barr is distorting the report and throwing shade at Mueller. To those supporting Trump, Barr is going out of his way to avoid criticizing Mueller because he is a close friend.

    As someone who is above politics and bias more generally, i would say that given the Special Counsel regs (28 C.F.R. § 600), Mueller's reasoning is quite peculiar.

RSS feed for comments on this post