Calth wrote:
Dyspeptic wrote:
I really wanted to avoid this thread, but the silver fox is not reductio ad absurdum, it's clearly a Straw Man. Or maybe False Equivalence:
Original argument: Class {Golem} normally has the attribute (literal name)
Counter argument: Class {Golem} is a subclass of Class {Creatures}
{Silver Fox} is a subset of class {Creature}
{Silver Fox} does not have attribute (literal name)
Therefore, Class {Golem} cannot have attribute (literal name)
Straw Man. Functionally equivalent to:
-----------------------------------------------------------------
Argument: Class {Mammal} has the attribute (hair)
Counter argument: Class {Mammal} is a subclass of Class {Animals}
{Fish} is a subset of class {Animals}
{Fish} does not have attribute (hair)
Therefore, Class {Mammal} cannot have attribute (hair)
Provably false.
Again that's wrong, because we know having a literal name cant be a rule for golems, as its not an actual rule for their type. You cant create an exception out of thin air. All creatures follow the same rules barring specific exceptions. If you want golems to have literal names, without citing an exception, all creatures must have literal names.
So to phrase it in your format:
Silver fox is a creature
Silver fox is not a literal name
Creatures therefore do not have literal names as an attribute
Class Golem is a subclass of creature
Class Golem has no exception listing literal name as a benefit
Therefore golems do not have literal name as a benefit.
Also, you chose a really bad counter example as fish scales and hair are effectively genetic parallels for each other (the genes that make scales in fish make hair in mammals), so nice job failing even there.
And arms and legs are genetically parallel to fins. Does that make arms and legs fins? No, no it does not.
In logic, all members of a class will share some attributes. That's what allows you to group them. Members of a particular subclass will have additional attributes that they WILL NOT share with the remainder of the class. You cannot generalize rules for one subclass from an instance of another.
Also, literally NO consistent system can be completely defined. There must, at some level, ALWAYS be underlying assumptions. The goal is to minimize this number, but they are always there.
And to parallel your attempted logic:
Silver Foxes are creatures.
Silver Foxes don't have opposable thumbs.
Class Humanoid is a subclass of creature
Class Animal has no exception listing opposable thumb as a benefit
Therefore humanoids do not have opposable thumbs as a benefit.
Oh wait,
Humans are creatures.
Humans have opposable thumbs.
Class Humanoid is a subclass of creature
Silver Foxes has no exception listing that they don't have opposable thumb as a benefit.
Therefore Silver Foxes have opposable thumbs as a benefit.
Hmmm... how is that possible? Hasty generalization, perhaps?
