The Dillahunty-Slick Debate: Can Chemicals Produce Logic?

Jeremiah Traeger

Jeremiah Traeger

I’ve had to slow down my blogging output recently, because graduate school has decided to make me drink responsibilities from a fire hose. A responsibility fire hose. I’m bad at metaphors.

However, in the past week, while I’ve been making figures dance on my computer screen, I’ve been listening to the debate between Matt Dillahunty and Matt Slick, hosted by the Bible and Beer Consortium, titled Is Secular Humanism superior to Christianity? I should emphasize that while that is the title of the debate, only Dillahunty appeared to want to have a discussion on the merits of Secular Humanism. Slick, on the other hand, decided that it was a trial of Glenn Beck-esque chalkboard free-association exercises to try and refute Secular Humanism by debunking naturalism instead. The debate went roughly as follows:

  1. Matt more or less rearranged his superiority of secular morality talk into the format of an opening argument, citing the foundation of human well-being as its source, and also citing its ability to change with new evidence as a strength.
  2. Slick tied Secular Humanism to philosophical naturalism (a position Dillahunty doesn’t hold), and then knocked it down utilizing many of the standard presuppositional apologetics, largely focusing on people not being able to trust physical evidence because they could be wrong.
  3. Whenever Dillahunty responded to Slick in a way that Slick appeared to not want to answer, Slick was able to dismiss it by giving the non-answer of, “that’s just your brain chemicals making you say that.”
  4. Repeat steps 2 and 3 for two hours.

You can watch the debate here:

For anyone who isn’t a fan of these types of debates (especially against presuppositionalists) I would recommend skipping it. This debate was a bit of a chore to listen to. However, this presented an opportunity to discuss some cool science shit, something I have neglected to do as NRR’s science expert.

Throughout the entire two hour, forty four minute debate, Slick appeared to only have one analogy in his analogy toolbox, which he brought up as a response to whether brains could do logic via purely naturalistic means. This response focused on brains functioning purely based on electrochemical signals in our neural networks, without any supernatural factors such as a soul weighing in. Slick repeatedly stated that a brain acting purely on physical mechanisms is like a vinegar and baking soda reaction, and that we could not gain “logical inference” from it.

Logical inference is a rigorous type of reasoning where the premises lead logically to its conclusion, synonymous with “deductive reasoning”. Simply put, you start with certain premises, which should lead to a certain conclusion. If the conclusion follows from the premises and the premises are true, then we can also state that the conclusion is true. However, if one of your premises aren’t true, then you can’t logically lead to your conclusion. For example, you establish that if A and B are true, then C is also true. But if A or B aren’t true, then you can’t infer that C is also true. For more concrete examples, check out this handy-dandy Wikipedia page.

Matt Slick brings up that purely physical reactions cannot use this reasoning, and therefore secular humanists “can’t account for logical inference”, while Christians can by inserting “god” in gaps as needed. Matt Dillahunty gave most of the reasons why this is a faulty argument against secular humanism, but I thought I’d use the opportunity go into how we can get reasoning out of purely physical processes.

In 2016 we rely on purely physical processes performing logical inference every day. If you are reading this on a screen, then you are relying on that process right now. Matt Slick relied on it through the whole debate as he took notes on his laptop. Circuits use logical inference every time we use them, as a result of simple inputs and outputs. Instead of “true” or “false” like the logic example above, integrated circuits rely on ON or OFF states in parts like transistors that your electronic device holds. If an electronic current is flowing through a transistor, then it is ON, which is a 1 in binary code. If there is no current then it is OFF, or a 0. Circuits can use these ones and zeros to perform all kinds of functions. In our logic example above, we required two true inputs to create a logically true output, which is analogous to an “AND gate” in a circuit. An AND gate has two inputs, and it will only output a 1 if both of the inputs are also 1. All the possibilities of inputs and the resulting outputs can be seen in the following truth table:



















If you look at the abstract logic example from above and submit 1 where you see “true” and 0 where you see “false”, you get these same inputs and outputs. This details how logic in circuits are directly analogous to abstract logical proofs. This is just one type of logical function that circuits can have, there are others too. There’s the OR gate, which requires just one of its inputs to be 1, or the XOR gate which requires either A or B to be 1 but not both. There are many other functions beyond just these.

Not only do purely physical processes perform logical functions, transistors are doing this constantly. If you’re reading this on a phone, you could be holding two billion transistors in your hand, which are constantly going on and off into their respective 0 and 1 states. Were physical processes unable to perform logical inference, none of our computers would work. Considering how integral computers are to our infrastructure and livelihood, if this type of functioning were to fail we would be what logicians refer to as “fucked”. Fortunately, circuits can perform this functionality.

Slick, as a former computer tech, knows how circuits work at a basic level under binary thinking. He establishes this in the Q&A when one audience member brings up a neural network. Of course, before the audience member even finishes his question, Slick interrupts him and insists that a computer model can’t show that a brain produces logical inference. The reason? Basically because it’s really damn hard to model the brain’s neural network and a brain and a circuit are not the same thing. Those statements are true, but then again, no analogy is perfect*. I’m really curious as to what the question actually was going to be. Slick seemed very insistent on shifting the conversation in the way that he wanted. I’m wondering if he would acknowledge that physical circuits produce logical inference or if he’d have some apologetic to explain that away as well.

Slick doesn’t state outright within the debate that physical processes can’t produce logical inference; he merely claims that chemicals cannot produce logical inference. Chemical processes are indeed a type of physical process, but there are physical processes that aren’t changes in chemical states (like circuits). When this audience member asks him whether physical processes can produce logic, he retreats to stating the following:

“One chemical state that leads to another chemical state… There is now way that has presented that we know of in any way, shape or form, how one state that leads to another chemical state produces proper logical inference.”

I find this interesting for two reasons, one because based on this answer he doesn’t rule out logical inference entirely from pure physics. I’m genuinely curious whether or not he thinks physical processes that aren’t chemical signals can produce logical inference. However, if he does think that circuits can produce logic, then he should have no problem accepting that chemicals can also produce the same thing.

Slick is right when he states that the brain is not the same thing as the model, particularly when comparing it to circuits. Circuits give their signals through the flow of electrons via wires, while the neurons in the brain go off of electrochemical signals like transferring calcium ions from one axon to another. However, in principle, they can do many of the same things. While, it’s true that neurons don’t give the same signal as a circuit, what’s important is that it gives a signal.

In fact, it’s simply not true that you can’t get logical functions out of changes in chemical states. Chemical state behaviors rely on inputs and outputs all the time. This is how we get things such as molecular circuits, molecular switches, and signaling pathways. At a very basic level, biochemicals undergo logical functions all the time, relying on certain inputs which create certain outputs. For example, a certain hormone will interact with a protein at a cell wall, which will release a signal comprised of something like ions into the environment, which will further come into contact with other proteins or receptors, releasing more signals, etc. Such behaviors are known as a signaling pathway, where the introduction of one chemical to the system may cause an output of a completely different chemical. A certain input gives a certain output, much like logic.


[Image: Signaling pathway in a cell, detailing the incredibly complex pathways chemicals go through in the cell to create certain outputs based on certain inputs. Source- Creative Commons by Roadnottaken]

In fact, there are people who are able to create genetic “circuits” that give logical outputs much in the same way that electronic circuits give logical outputs. Christopher Voigt is a researcher at MIT who has essentially created a programming language for cells, titled Cello software. A researcher is able to design a simple genetic circuit that has similar logical gates discussed above, but instead of wires, the genes output chemicals through a particular pathway of gates until there’s a final output. The paper which discusses this (which is unfortunately behind a paywall) states that 37 circuits that they designed gave a clear ON or OFF output as a result of a desired input, meaning that they were following through logical processes to produce a desired state. To put it in Matt Slick’s terms, there was a difference between one chemical state and another, which was achieved through the use of a logical process.

Let’s look at the brain then, keeping in mind that I’m far from an expert in neurology. We know the basic functioning of how neurons work, transferring an electrochemical signal from one neuron to the other, which can change the total state of the brain when neurons in bulk give signals. We know that the brain receives many inputs from neural pathways that reach across the body and multiple sensory organs, as well as giving outputs to other organs. Is it too much of a stretch to think that the brain can receive inputs like words and numbers from sounds and visuals, translate them through a series of complicated neuron signals, and create a brain state that produces a correct answer? Given that we can do that for simple circuits and molecular reactions, it’s not exactly far-fetched to conclude that a complicated organ such as the brain can do the same thing. It doesn’t have to be absolutely perfect, as no physical process is, but absolute certainty doesn’t appear to be attainable for most beliefs anyway.**

We don’t know a lot about the brain, still. Even in 2016 neurology seems like a wide-open frontier, and there’s a lot to discover. Is consciousness just a certain pattern of signals that we perceive as our own identity? How do we store memories? Could we “solve” the brain, such that we could read someone’s mind or make our own? There’s a lot to say about it, but one thing we know for sure is that it’s complex, and we have far more to learn. When Slick interrupted the questioner, he made sure to push his answer towards emphasizing that the brain had been designed and it’s far too difficult to model, at least for now. From my perspective, maybe there’s some mystical soul, a ghost in the machine, driving our bodies to move, but I see absolutely no reason to accept that as true. While I don’t claim to have perfect knowledge of how brains work, it seems perfectly reasonable that they could perform amazing, though imperfect, logical tasks, especially considering its complexity. Perhaps Slick is content in saying that our brains can never perform logical inference, but I’m not. That seems like an unjustified claim. This uncertainty means that we have to put more work into figuring it out, and a mere appeal to the supernatural doesn’t do anything but make us curtail our attempts to understand it. Our brain is far more than fizz, so let’s investigate what it is!

For the record, Slick poo-pooed the questioner for drawing an analogy between a brain and the model comprised of circuits, saying that since we can’t model the brain and it’s so complex that it would be silly to compare the two. I find this incredibly dishonest coming from a man who spent almost all of his speaking time comparing the brain to a simple reaction between vinegar and baking soda. He does not get to make a ridiculously reductive comparison, and then shame someone for someone making something that is far closer in design, even though it’s imperfect. For this reason I found his treatment of that questioner incredibly dishonest.

So, to sum up:

  • Simple processes like electron currents in a circuit can create logical inference.
  • Chemical reactions can change chemical states in a similar way, going through a logical pathway much like we treat logical arguments.
  • Brains are comprised of a network of neurons, which transfer signals to each other from cell to cell through simple electrochemical processes that give rise to much more complex behavior.
  • It would be special pleading to say that neurons can’t send simple chemical signals to each other in a way that other chemicals can.
  • We still have a lot to learn, and a simple statement like “the brain can’t work like that” is unsatisfactory.

These are the lessons for this blog post. The lessons from the debate are entirely separate. I would hope, though, that the biggest lesson that Slick took away from this debate is that if he’s going to argue against secular humanism, he’d better stick to the topic he signed up for if he wants to be taken seriously.

Edit: Matt Dillahunty responded to me on Twitter regarding my points.

@Matt_Dillahunty: That wasn’t Slick’s point. This is about whether there’s a solid, objective foundation for the reliability of reason

@nrrprophet: Not saying it was the foundation of his arguments, but he made claims that chemicals couldn’t create logical inference, no?

@Matt_Dillahunty: Not at all. He’s pointing out that a materialist worldview can’t ever move beyond the brain to justify reason.

@nrrprophet: I quoted him stating that chemical rxns can’t produce logical inference. I really value your feedback though.

@nrrprophet: if he thinks that circuits can’t produce logical inference, then my points here are moot except as a science lesson

@Matt_Dillahunty: may have been a slip… Because that’s not his objection

It may have been a slip up, but he raised it multiple times. Dillahunty addressed the meat of the issues they were debating, I have no reason to go into that further. The point of this post is to address whether or not physical processes can perform inductive logic, and Slick claims that they can’t. The logical absolutes are irrelevant to my point here, but if they are the foundation of Slick’s arguments, then I haven’t addressed them here. Hopefully this is informative to readers, though.


*As my partner-in-crime Ari has stated once, the only perfect analogy is a tautology. That is, the only time an analogy will not break down is when you are comparing something with itself, which is not particularly useful.

**It’s worth noting that in this debate, Matt Slick claimed not to be appealing to absolute certainty. This is wise, as Dillahunty has made it clear at his debate with Sye Ten Bruggencate that he doesn’t care about absolute certainty. In real physical processes, circuits fail, molecules decompose, and in a reaction the chemicals will never be used up 100%. That’s ok. We can still discuss the logic being formed recognizing that sometimes you will get the wrong logical output due to occasional failure. As long as we look at the inputs and functions of the logical process, we can determine its most likely output, recognizing that we won’t be absolutely certain.


Tagged: , , , , , , , , , , ,

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: