PHL312 Philosophy of Computers and Computation
Professor: Craig DeLancey
Hi, sorry, I have to move my office hours a little bit.
I hope that's OK. Today they'll be 10:30 am till 12:00 pm.
Here's the link:
Read the section of the paper "Analog and Analog" called "Digital."
This is pages 213-219. The section "analog" is recommended but
optional. The paper is on BlackBoard.
Class will be online. Here is the link:
We'll discuss TMs a bit more, and look forward to
our first puzzle (semantics).
Before class, please read "The Chinese Room" paper (which is
actually titled "Minds, Brains, and Programs.") It is on BlackBoard.
Read at least the first 5 pages before class; if you're swamped you
can finish the rest before Wednesday (the whole paper is just 13
pages long--the rest is notes etc.)
Apologies! I have conflicts today and will have to move my
office hours to Wednesday (which will then at least be in
person!). I hope that's OK! Email me if you want to talk today,
though, and I can make some time between appointments!
Before class, finish Searle's "Chinese Room" paper.
We'll discuss the Chinese Room Thought Experiment some more, and
we may peek ahead to the frame problem.
Please take a look at Tursi. On
jar download, you should be able to download a copy of the program.
Please do see if it will run on your preferred
Office Hours: I will have office hours in my office from
3:00 - 4:30. First in-person hours! Sorry I have too many meetings.
Office hours 9-11 are at:
Please read "Cognitive Wheels" by Dan Dennett. It's on BlackBoard.
I'll post some study questions for you later today.
Reading: read parts 1-3 of Turing's "Computing Machinery
and Intelligence." Please read part 6 before Wednesday, but
it'll be great if you read it before Monday. A version is
Parts 4 and 5 are optional. I have also put a copy on BlackBoard.
What does Turing mean by "the imitation
game"? Be able to describe it. Here are some other question to think
on while you read:
- Why does Turing want to avoid trying to define
intelligence? What challenges do you think that there might be to
- Describe the imitation game (now called the Turing Test).
- What does Turing mean by "machine"?
If you cannot run Tursi, you could write the Turing machine
"in your head." Tursi helps us to test program, and can prevent
you from making some mistakes, and it lets you "watch" a turing
machine running; but it should be possible to write the program
as we did in class. In such a case, if it is more convenient for
you, you could write the machine on paper and hand the paper in
Homework: making two Turing machines. You may work in
teams of 3 or fewer people. What you hand in will be two (or
three) text files. In the comments section, include the team
members and also how the tape must be prepared. (Some people
are confused about this. A text file is a kind of file--Word
files and rtf files will not open in Tursi. In Word, choose
"save as" and then choose "text" to make a text file. If you
are confused, email me.)
Thus, your text file can begin with something like:
# My team includes: Craig DeLancey
The first five lines are comment lines. They tell who is on
the team and also how the tape must be prepared and how to find
the answer. The sixth line specifies that the start state will
be 0. The last says that the halting state will be H. You can
of course choose your own symbols for start and halt states.
Below all that, you list the rules.
# This addition machine adds two numbers represented in unary
# as a series of 1s, separated by a single space. The machine
# starts on the leftmost 1 of the left number. The machine ends
# under the the rightmost 1 of the answer number.
#! start 0
#! end H
You are only handing in this text file; or, if you do the program
by hand, then you can just hand in your code written out on a page.
The Tursi program is
just to test your program. You don't write inside the Tursi
program--except to put something on the tape that you can test.
(Here is the simple example
we made together in class. Look at it for a model.)
The machines you make will be:
EMAIL your files to me. That way, I can load them up myself.
Please name your text file "LASTNAME.1.txt" and "LASTNAME.2.txt".
If you are a team, just pick one last name. I just need to be
able to keep them separate from other people's homework, so this
- a machine to tell if any given number is evenly divisible by three.
- an addition machine that can handle zero (n+0 or 0+n)
- Extra credit for the maniacs among you: write a multiplication
I was asked for some advice on a few things.
- It's best if the answer is on the tape (as opposed to saying
the answer is in the state of the machine itself). You can say
how the answer is put on the tape.
- You can use whatever tape alphabet you want! You are not
limited to 0 and 1!
- A common error is to give the machine conflicting
messages. This will be rejected by Tursi. The whole
point of the machine having internal states is so that you
can do different things when handling the same input.
Suppose sometimes you want the machine to write 1 when
reading 0, and sometimes sometimes you want the machine to
write 0 when reading 0. You cannot say:
1 1 1 R 1
That's contradictory. Consider instead:
1 1 0 R 1
1 1 1 R 1
2 1 0 R 1
Class is online today. Here's the link for class:
Office hours online from 3:00 - 4:30 pm, here's the link:
Please read before class: Sprevak's
review of the "triviality" arguments before class. We'll discuss
this in the next several classes.
Zach sent a link to this video:
We're Building Computers Wrong (for artificial intelligence)
. The question that our critics raised was: can you
send arbitrary messages efficiently in this way? Can you
control errors with an analog coding? (Remember our case of:
counting a deck of cards versus measuring the height of a
deck of cards.)
Office hours 2-4 pm at:
Complete and hand in the simple handout of a homework on DFAs that
I gave you.
For those of you still looking for a Turing Machine simulator,
here is one that seems to work in most browsers. It orders
the table columns differently but is otherwise the same (that's not
a significant change.) Play with it to get an idea of how the Turing
Reading: Before class, read the short selection from Kripke, available now on BlackBoard.
Read pages 6-22. Answer the questions on BlackBoard about the reading.
We are turning now to our last philosophical puzzles:
We'll have a few fun readings for each of these.
- Philosophical Puzzle 4: what counts as an algorithm? Do algorithms exist independently of their implementations? Can algorithms act on the world?
The Kripkenstein paradox for machines. The paradox of infinite cases. The argument that algorithms are irreducible
- Philosophical Puzzle 5: Information again; Vitalist problem; Knowledge as independent information; memes.
- Philosophical Puzzle 6: The simulation argument. Are we living in a simulation?
- Philosophical Puzzle 7: Are simulations “real”?
- Philosophical Puzzle 8: Are you a bad person if you do bad things in a simulation?
- Philosophical Puzzle 9: Is computer science a science?
Office hours today must start and end an hour earlier. Sorry for the
inconvenience. Here is the link:
Reading: Before class, read the short selection from Kripke, available now on BlackBoard.
Read pages 32-44. Recommended is 22-32 and also 44-54, but I know you are busy!
Assignment: Short paper (3-4 pages, double spaced, 1" margins) due on one of the following
Please use my
Analytic Philosophy Paper format.
- Is Searle right that a machine with only syntax at its disposal
cannot manage to have meaning? Very briefly explain his thought experiment.
If you disagree with him, what error does he make? (For example, is meaning just syntax?
Or can we make meaning with syntax alone?) If he's right, what do you think
this meaning thing is such that the isolated algorithm running lacks it?
- Is the Frame Problem solved by meaning? That is, are we able to avoid
(superficially, at least) the Frame Problem by having these things meanings
at our disposal? How does that work? If we don't have them, then what if
anything seems to distinguish us from the machines falling to the problem?
- What is the solution to the "every rock is running PowerPoint" problem?
Is it a problem (that is, maybe it's ok that every rock is running PowerPoint)?
What is your solution to the problem?
Recommended watch: before reading the Hofstadter, it will be helpful to watch this goofy fun video:
Reading: before class read the selection from Hofstadter. This is available on BlackBoard.
Practice: Answer the question on BlackBoard about the reading.
This may seem unrelated to our Kripkenstein reading. However, note that the question here is: are
algorithms things that have effects in the world, that we need to explain the world, and therefore
perhaps about which we should be realists? In a funny way, this is the mirror image of the Kripkenstein
problem. It also is essential to a question we will ask later: is computer science a science? Or is it
something that could be replace by physics, only we're too lazy to do that?
I'll have office hours in my office (MCC212A) from 11:30 am -2:30 pm.
1, 4 April
Reading: before class read the simulation argument by Bostrom.
Practice:: answer the questions on BlackBoard about the Bostrom paper,
meant to help you think about its argument.
A note about grading. We originally had in the syllabus that we would have a
midterm and a final. Since we're considering making this a writing class, I think
the two short papers can count as our primary assignments, but we can have
a short final where I give you the questions before-hand in order to study.
Office hours 9:00-11:00 am at
Quest. There's a philosophy session from 11:00 - 12:00 in MCC210. Two members
of our class are speaking, why not join us for an hour!
I will have in-person office house in MCC212A from 1-3 pm. This will be instead of
my Thursday office hours.
Reading: before class read the selection from Chalmers.
My office hours need to be later today, sorry. They're 11:00 am -
12:30 pm at:
We'll continue our discussion of: are simulations real? And, we may turn to:
does what happens in a simulation matter?
Office hours 2-4 at: https://meet.google.com/uza-dokd-hic?authuser=0&pli=1.
The philosophy club will meet at 5:30pm to watch an episode of Black Mirror
called "Nosedive" in MCC room 242. The episode focuses on what social media and the quest
for popularity and social acceptance can do to people. Room TBA.
I'll be in Arizona giving a talk and attending talks at a conference.
We'll have some asynchronous online classes this week! Our themes will be:
does what happens in a simulation matter? Watch my online lecture on
a basic question regarding simulations. Then, complete one (just one) of the following
tasks; answer the one question on BlackBoard about the reading or film you chose
(I list the questions also below).
- (This one is a little disturbing; skip it if you want to avoid some
disturbing content.) Read The Nether, which is on BlackBoard.
Did the Nether do a kind of harm?
If so, what kind of harm? Explain it. Should the Nether have been shut down?
Why? Be very careful to clarify the status of the harm (if any) that was
committed; what kinds of things were harmed, for example, if anything?
- Watch The Matrix, if you haven't done so and if you have access
to a copy (the library used to have one but seems to have lost it; sorry).
Consider the scene from 1:41:00 to 1:44:15; this scene is
here. What did
Neo and Trinity just do? Was it morally justified? Explain why or why not.
Be very careful to clarify the status of the harm (if any) that was committed;
what kinds of things were harmed, for example, if anything?
- Watch Ready Player One. The library has a copy at
PN1997.2 .R43 2018, you can get headphones and watch it there, if you've
not see it. The protagonist Wade has two lives or locations: online, and the
actual world. It looks like Wade is going to spend much of his life (at least
5/7ths!) in the
simulation. Is this a mistake? Is he wasting this portion of his life because the simulation
doesn't matter? Or do his life and his accomplishments in the simulation
have as much value as his life and accomplishments in the actual world? What
if he spent all his time in the simulation? Would that be a good life?
In class, we will do 3 things:
I'll have office hours in my office MCC212A from 3:00 -- 4:00 pm.
- Discuss your thoughts from our assignment; can virtual environments matter?
- I think we should review the Halting Problem, and I'll show you Deutsch's proof that there exist CantGoTu environments.
- We can begin our discussion of memes, perhaps with a little historical discussion of vitalism and genocentrism.
Office hours online 9-11 am at:
Reading: before class read the selection from Dawkins. This is available on BlackBoard.
Very Unserious Assignment: bring an example of your favorite meme to class.
Reading: I'll bring you a handout about the Popperian view.
Read the first four pages of the Clark and Chalmers handout. We'll
discuss: can computation extend beyond the computer? Then we'll turn
to our last question: is computer science a science?
We'll continue with our last topic: is computer science a science?
Then we can review the entire class in a breathless rush.
Recommended reading: on Blackboard are a few pages from
Rapaport on the nature of science and computer science. I recommend
you read them before class--it's very straightfoward, easy read, like
a well-written news article almost!