If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:

Or download the MP3


It’s old news by now Hillary Clinton had a pretty rough time of it on September 11th. I’m sure you’ve heard the story and you might even be sick of it, so I won’t spend too much space on the details. I’m more interested in looking at it from a slightly higher level than that. In particular I’m interested in an examination of figuring out how to figure out what’s going on. Meta-figuring out you might say. Or perhaps meta-investigation sounds better.

In my reading I end up in some of the darker corners of the internet. And the idea that Clinton might have health issues has been floating in those corners for awhile. Of course if some person, in some dark corner of the internet says that Clinton is sick that doesn’t mean anything, but it doesn’t mean nothing either. However partisan someone is, they generally have to have something to latch onto first. In this case it was Clinton’s 2012 fainting spell where she hit her head and got a concussion, and when later examined was found to have a blood clot.

On its face that sounds serious, though, in the interest of full disclosure I have known two people who died as the result of, or from complications due to, blood clots, and both were comparatively young, so I may be predisposed to view clots as more dangerous than they actually are. Even so, I wouldn’t say I decided anything right then, certainly not that Clinton definitely has a health problem and there’s a conspiracy to hide it. Rather it was one of those things that you file in the “might be true” box which also puts it into the potential black swan category. Of course given the enormous power of the president and the fact that they’re only one person a president’s health is always a potential black swan, just look at William Henry Harrison. But when you hear something like that you might, without making any firm decisions, mentally increase the odds a little bit.

That’s where things remained for a while and that’s where they might still be, had no further facts come to light. But of course more facts did come to light. Also, the election is less than two months away, so if I am going to do something with it, I need to do it soon.

Now we’ve already talked about how any one person is very unlikely to influence the election, so if the only thing at stake is my vote, then who cares what I end up deciding as far as Clinton’s health? But of course it’s not just me. There are thousands, if not millions of people out there who have all heard the news about Clinton’s collapse, and are now trying to decide if Clinton’s health should be a factor in how they cast their vote. How do these people make up their mind? Where can they go to get their information?

I said earlier that more facts have become available, and while that’s true, facts are not what most people have access to. There are facts and there is the spin on those facts. And mostly what’s available is spin, which you then have to dissect to get the actual facts. And this is where the difficulty arises. This is where meta-investigation comes into play.

I’m going to take you through a few of my own attempts to do this dissection, mostly to illustrate the difficulty inherent in the process. Is this going to be horribly prejudiced? Almost certainly, but I think despite that, viewing the process might be valuable anyway. I am going to try to keep it as neutral and objective as possible, but just by choosing the subject of Hillary’s health, people are going to claim that I’ve clearly picked a side. Perhaps I have, though, honestly, I’m almost certainly going to vote third party. But certainly, by focusing on Clinton’s health, people who are looking for reasons to disqualify my opinion would have ample excuse right there. But if you’re looking for an excuse not to listen, I think you may already be dealing with an significant lack of objectivity without any help from me. And regardless of your political leanings I think the issue of how to get at the truth is important, and Clinton’s recent health problems just seem to be custom made for this sort of thing.

Returning to Clinton’s most recent episode on 9/11, as I said, obviously at this point there is already tons of stuff out there from all sides trying to spin the event as either not worth talking about at all, (which seems a stretch) or the end of the Clinton campaign (also a stretch). With such an enormous amount already written on the subject where does one go? How do we extract anything?

For myself I started with the video. What can we get out of that? First no one seems to be claiming that it’s not Clinton, which is a great start. It’s amazing how once you get into the weeds what sort of theories actually get floated. But it can be necessary to consider even the crazy stuff if you’re really trying to strip the facts from the spin. But it seems safe to put the video in the facts column. The video is an actual record of an event which actually happened.

Okay, so that’s a fact. What do we do with it then? Well on the one hand you’re not a doctor and you weren’t there, but on the other hand you do have a decision to make on November 8th, when it comes time to cast your vote. And, hopefully, you want to make the best decision possible, so while viewing a video isn’t ideal, it is information largely without a filter. What other facts can we extract?

Unfortunately not many. But wait, you may be saying, what about the fact that she was just overheated? Or that her doctor said that she had pneumonia? Or if we’re talking about videos being facts, what about the video a short while later when she emerged from the apartment and appeared to be doing great?

Let’s take those in reverse order. There’s a quote that I really like, that I think cuts to the heart of separating fact from spin:

Journalism is printing what someone else does not want printed; everything else is public relations.

The origin of that quote is obscure. But it’s basically a restatement of a principle often used when attempting to separate historical truth from fiction: the criterion of embarrassment. The idea is that if you see something people don’t want you to see it’s probably true and when you see something that someone wants you to see it’s more likely to be staged or spun for your benefit.

Applying this to the two videos. It’s very unlikely that Clinton wanted anyone to see her having a hard time walking to the car. In fact it’s been reported that she forbade cameras from filming her. And the only reason we have a video is that a private individual filmed it on their phone. Thus by the criteria of embarrassment this is probably a true glimpse of what was going on. When you look at the second video it was obvious that she wanted cameras to be filming and wanted people to be watching. Does this mean it wasn’t actually happening? No, but it means that she (or her handlers) decided that this was something that should be seen. So while it was undoubtedly true, it’s also possible that it had less information content than the other video. Still if you want to enter it into the record as a fact you may, but it should be entered with the intentionality of the act understood. Clinton decided to appear in front of the media later that day.

What then, about the overheating and the pneumonia? This relates back to the criteria of embarrassment. Clinton’s doctor and staff is telling you what they feel comfortable telling you. There is certainly some embarrassment involved but less so than not saying anything. And the fact that there was some lag in mentioning the pneumonia means that at some point they did not feel comfortable mentioning it.

Finally the heat exhaustion and the pneumonia raise their own interesting points. If we turn again to trying to get at some facts is there anything to be extracted from those claims or more specifically the timing of those claims

I find the statements on pneumonia and overheating to be uninteresting when considered separately. It’s only when you consider them together that I something potentially useful emerges, but first we have to step back a bit. Clinton’s doctor says that she was diagnosed with pneumonia on Friday. And since the video of her collapse was posted, both Clinton and her staff have apologized for not mentioning the pneumonia diagnosis sooner. Obviously if she was diagnosed on Friday, then Clinton was aware of it, but it also sounds like some members of her staff were aware of it as well. If everyone was aware of it, when the video and other news of her difficulties on 9/11 came out, why didn’t they immediately say “She has pneumonia.” Instead, in the immediate aftermath the story was that she had merely overheated.

I understand that when someone get’s this nitpicky they are in danger of starting to sound like a conspiracy theorist. But when you’re trying to extract facts from spin, dissections like this can be pretty valuable. And what we might be seeing here is one of those instances where you get a view of some facts before the spin machine has a chance to take over. Also as Taleb said a few days after the incident:

It is as irrational to reject all conspiracy theories as it is to accept them.

With that in mind, let’s get a little bit nitpicky. As I said, in the immediate aftermath of Clinton’s troubles and the video, no one offered up the information that she had pneumonia. But at this point a few days out, most people probably don’t even remember that pneumonia didn’t come up till later, or if they do they assume that the pneumonia diagnosis was a clarification which followed quickly after the explanation that she overheated. In reality it was many hours after the fact that they announced that she had pneumonia. Clinton collapsed at around 9:30 that morning. They announced she was overheated at 11:00. She appeared in public in the second video around 1:00. The pneumonia didn’t come up until 5:00. (This is a good, if partisan, breakdown of the timeline, but the NYT confirms the key fact about the pneumonia not coming out until 5:00.) This timeline takes us back into the realm of facts. But once again we’re left with the question of what to do with these facts.

As an aside, it’s perfectly fine if you’ve decide that Hillary Clinton could be in a vegetative state and you would still vote for her over Trump. But if, for whatever reason, you have decided that Clinton’s health is a factor in your decision on how to vote, perhaps even the deciding factor, then trying to get to the bottom of what’s going on is important. And even if her health isn’t the deciding factor, and something else is, you will still face the same challenges getting to the bottom of that issue. Particularly given the incredibly polarized nature of today’s media and politics.

But returning to the timeline. If lots of people knew that Clinton had pneumonia as early as Friday, why wasn’t that the first explanation offered? I understand that’s a question which is impossible to answer, just as viewing the video doesn’t definitely answer the question of whether Clinton is healthy enough to be president. So we’ve managed to tunnel down to the level of facts, but having reached there we may be no closer to a decision than we were before, and we probably have no choice but to wade back into the “spin zone.” However at least we have some facts, so that when we do wade back in we’re better prepared to decide which spin might actually contain an element of truth. Presumably we’ll be particularly attentive to that spin which accepts the same facts as we do, and we should reject out of hand any spin that denies the few things which we’re certain are factual.

Of course once you’re back into the spin zone it becomes even more difficult. For example the first thing you might want to do is get a doctor’s opinion. This is a health issue after all. Obviously there’s Clinton’s doctor, and whatever they say shouldn’t be discounted. But under the criteria of embarrassment you may nevertheless decide to set it aside, or even if you don’t it’s perfectly understandable to seek a second opinion. This is another point where people might stop reading, since if you choose to discount Clinton’s doctor then you’re, by definition, putting forth a conspiracy theory, the theory of a conspiracy between Clinton and her doctor (and probably others as well) to lie to the public. First, refer back to what Taleb said, second it’s not as if conspiracies to cover up presidential or presidential candidate’s health are without precedent. But as I said, maybe you just want another perspective. Certainly if you looked around and all doctors were unanimous in declaring that the symptoms Clinton had exhibited (going all the way back to the 2012 fainting episode, and perhaps even farther) were perfectly normal, then that would be a reasonably good sign that these health scares were overblown. What if, on the other hand, you found some doctors who were worried? What do you do then?

This is where the wading into the spin becomes so problematic. If you’ve done what you could to uncover facts, you’ve watched the video, and it doesn’t raise any concerns. You’ve uncovered the large gap between the incident and the first mention of pneumonia and you feel like you have a reasonable explanation for that. Then you might decide to completely ignore the doctors who have expressed concerns. And if that’s the decision you’ve reached, that’s completely reasonable. But what if, having seen the video and uncovered the time disparity you have a nagging feeling that it doesn’t add up. And then you come across a doctor, presumably more knowledgeable than you, who has these same misgivings? Should this be added into your own assessment of probabilities? Or should you ignore it because this doctor obviously hasn’t done an actual examination of Clinton? Being human it’s certainly going to be the first, but still you might want to look closely at the doctor, does he have any biases you should be aware of?

This exact thing happened to me, albeit before the incident on 9/11. Drew Pinsky, often known as Dr. Drew, who for many years was on Loveline with Adam Carolla came out with his misgivings about Clinton’s health. Okay so there’s a doctor, and he also thinks something might be going on there. (Once again to be clear I am not basing my vote on my assessment of Clinton’s health, I just think it’s an interesting exercise in getting at the truth. Whatever that is.) But what do I know about Dr. Drew? Obviously I’ve heard of him. I saw a couple of episodes of Loveline, several months ago I listened to his interview on WTF. But beyond that I don’t know. Is he a notorious conservative who will say anything? Does he have a long standing feud with the Clintons? It can be hard to tell after the fact because of course the minute he offers his opinion, not only is that opinion tossed into the partisan battlefield with one side rallying around it and the other side attempting to blow it up, but Dr. Drew himself is forced onto the battlefield and forced to take sides. He may have been completely apolitical before this, but suddenly if he doesn’t want to be completely alone against a fairly furious attack he has to pick some allies.

As you can see anytime someone does offer a definitive opinion, even if backed by expertise, it get’s swallowed into the partisan maw. So where does that leave someone who’s still trying to figure out the truth about Clinton’s health? To further complicate things, it should be pointed out that neither side wants you to succeed. Even if one side is more correct than the other they are both trying to push the pendulum as far as they can. One side want’s to convince you that there’s a 95% chance Clinton will be dead by Election Day and the other side wants to convince you that Clinton had the mildest form of pneumonia possible, barely worthy of the name and other than that she has the stamina and intellect of a 25 year old.

I wouldn’t blame you if, after all this, you threw up your hands and decided to give up on the whole enterprise, or if, as is far more common, you picked a side at some point in the past and decided to just believe whatever that side was saying.

But if you’re still determined to dig. And once again recall that your vote, particularly in the presidential race, almost certainly doesn’t matter. So that even if you do arrive at a firm, unshakable position, that you’re basically King Canute commanding the waves. If despite this, you still want to continue, then I offer my final piece of advice.

One of the things that you’re naturally going to be inclined to do, and which you’re going to be pressured to do, is make a firm and final decision. In the case we’ve been examining you’re going to be asked to declare once and for all that Clinton is hiding something, or alternatively that she is not. In fact people are going to want you to go even farther and not merely declare she is hiding something, but toss in decades of misdeeds by the Clinton’s as well, or alternatively, demand that in addition to vouching for Clinton’s transparency about her health, that you add in a declaration that Trump is a misogynistic moron. You should resist this impulse and this pressure. Nothing is certain. What you should really be doing is adjusting your probabilities, not trying to find some firm and final answer. This doesn’t carry the certainty of deciding that everything is fine and there’s no reason not to vote for Clinton. But remember that you can’t command the waves, and contrary to how the story of King Canute is normally used these days. The King knew that, and the point of the whole exercise was for him to show the limits of his power, just as we need to be aware of the limits of knowledge.

With this in mind let’s return to an examination of the recent comments by Dr. Drew. If you dig into things there are some elements which move the needle in one direction and some elements which move it in the other direction. As an example of something which makes Dr. Drew less credible you have to of course include that he evaluated her without access to her records and without any kind of examination. Not only is the information he’s relying on partial and potentially misleading, but it’s a borderline violation of medical ethics.

As an example of something which adds to his credibility, if you look at what he actually said, he’s not only commenting on Clinton’s health but her medical care. When he says that some of the medications she’s receiving are no longer recommended, then not only is that a statement devoid of politics, but it’s also very specific and detailed, something you could probably check if you were so inclined.

Obviously there’s a lot more than just these two, but in the absence of a smoking gun, deciding on a probability and adjusting it as new information comes in, is the best anyone can do, particularly with the incredibly low signal to noise ratio of today’s journalism

I fear, having reached the end that perhaps this isn’t as useful as I hoped. But of course there are no easy tactics to uncover the facts. Still, hopefully the criteria of embarrassment was something useful to add to your cognitive toolkit. And finally, remember:

Journalism is printing what someone else does not want printed; everything else is public relations.