Introduction to Philosophy of Science Unit 4: Explanations, Theories, and Models
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Recommend
9 
 Thumb up
 tip
 Hide
Some puzzles remain from last time, such as what to make of the fact that some sciences, such as physics, make use of apparently incompatible theories or paradigms without any of the scientists involved seeing this as in any way inconsistent or horribly bad. In this unit, we will look at one way of making sense of this, among other things. We will look at what exactly a scientific explanation is supposed to be as well as what theories are in the first place. This will lead us to our final view of science - that of Ronald Giere.

READING
Theory and Reality Chs. 12.7 and 13
RPS selection 12: Carl Hempel, "Laws and Their Roles in Scientific Explanations"; selection 13: Wesley Salmon, "Why Ask 'Why?'?"; selection 14: Bas van Fraassen, "The Pragmatics of Explanation"; selection 15: Philip Kitcher, "Explanatory Unification"
Ronald Giere, Explaining Science: A Cognitive Approach Ch. 3

SUBSCRIPTION THREAD FOR THIS SERIES
http://www.boardgamegeek.com/thread/494877/introduction-to-t...
Your Tags: Add tags
Popular Tags: [View All]
  • [+] Dice rolls
1. Board Game: Sensations [Average Rating:3.17 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Theories as Explanations

Common idea:
 Science deals with theories
 Science offers explanations

Plausible connection: Science offers explanations by giving us theories
But what sorts of things are theories or explanations?

The Positivist View of Theories

A theory, according to the positivists, is a certain kind of set of sentences
Other than purely logical or mathematical vocabulary, theories could contain two basic kinds of terms
 Observational terms
 Theoretical terms

Observational Terms

Observational terms are those whose application can be determined "just by looking" - that is, in a very immediate way, without having to go through any process of reasoning
Observational terms are meaningful because of their direct connection with experience

The positivists disagreed on whether terms referring to ordinary objects (tables, cats, etc.) could count as observational terms or whether instead only terms referring to sensations should count
If only sensation-terms count, most of the non-logical terms in our language would fail to be observational
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
2. Board Game: _Word_Bridge_ [Average Rating:6.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Theoretical Terms

Theoretical terms are those meaningful (non-logical) terms which are not observational
But how do they get their meaning?
This is the same problem we saw earlier in Unit 1:
 Many scientific theories appear to refer to entities which might perhaps be incapable of being directly observed
 The general strategy:
---The general positivist attitude was that talk about, say, atoms was really just shorthand for talk about things we can experience directly
--- Reductive behaviorism: Talk of mental or psychological states is really just talk about behavior

Theoretical terms get their meaning from their connections with observational terms
These connections come in the form of definitions or bridge laws (also known as correspondence rules) which define theoretical terms using purely logical and observational terms

Later, these bridge laws were no longer seen as definitions but as coordinating theoretical terms with observation in important ways
It was still seen as necessary to connect theoretical terms with experience in order to give them meaning
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
3. Board Game: SET [Average Rating:6.48 Overall Rank:1229]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
The Positivist View of Theories

A theory, then, is a set of sentences including all and only the following:
 Laws of logic/mathematics
 Applicable bridge laws
 Theoretical postulates
 Anything deducible from these
(A theoretical postulate is simply a factual statement of a scientific hypothesis)

Criticism

There is no clear distinction between observational and theoretical terms
What is observable by one person may not be so by another
There is no sharp cut-off between what is observable and what is not
Observation itself is theory-laden
Restricting observation to sensations alone suffers from various criticisms
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
4. Board Game: Dice Games Properly Explained [Average Rating:6.65 Unranked] [Average Rating:6.65 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Explanations

Two parts to every explanation:
 Explanans: the thing doing the explaining
 Explanandum: the thing being explained

Initially, it is plausible to think of an explanation as a set of sentences, with one being the explanandum and the rest together being the explanans
Already, explanations appear similar to arguments, which also feature two parts - premises, and conclusion

Hempel on Explanations

Positivists wanted to develop a view of explanation from this apparent insight
Two requirements for a good theory of explanation:
 Explanatory relevance: Explanation gives us good reason to believe the phenomena explained does, did, or will occur
 Requirement of testability: Statements composing an explanation must be capable of being subjected to empirical test

Hempel thinks his view meets both requirements
An explanation is simply a good argument of a certain sort, with the explanandum as the conclusion
Since it is a good argument, it will give us good reason to believe the explanandum
Hence, explanatory relevance is satisfied
Since arguments contain only statements that can be true or false and only
meaningful statements can be true or false, it follows from the positivists’ verificationism that arguments contain only logical truths or verifiable truths
Hence, the requirement of testability is also satisfied
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
5. Board Game: There Oughta Be a Law [Average Rating:0.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Generalizations

A candidate for an explanation must contain at least one law-like generalization
What is a law-like generalization?
A law-like generalization is a generalization which, if true, would be a law

Contrast this with accidental generalizations - generalizations which, if true, are only accidentally true
 Ex.: All red-heads are left-handed

A key difference between law-like and accidental generalizations is that law-like generalizations - but not accidental ones - support counterfactuals
A counterfactual is just a claim about what would have been the case in certain possible but nonactual circumstances
A true law-like generalization will ground or contribute to the truth of certain counterfactuals
 Example: The more a gas is heated, the more it expands.
--- Suppose this is a law-like generalization
--- If it were true, it would also make counterfactuals such as the following true:
------ If I were to take a gas at 54 degrees Fahrenheit and heat it to 63 degrees, it would expand.
This is a key difference, but the nature of the distinction is a subject of controversy

The Covering-Law View

(Also called the ‘D-N’ or ‘Deductive-Nomological’ view)

An explanation is an argument such that:
 The conclusion is the explanandum
 At least one law is included among the premises
 All premises occur essentially
 All premises are true
 The premises give us good reason to believe the conclusion
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
6. Board Game: Form A Word [Average Rating:4.69 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
The Symmetry Thesis

Hempel: Explanation and prediction have the same form
The only difference is whether we start with the conclusion or with the premises

Example:
This gas sample is confined to a fixed volume of 2
I have just heated it to 100
T=PV
So, the pressure of the gas is 50

Often when we give explanations, though, it doesn’t always seem like we invoke a law or all of these various premises
Hempel: This is not because these things are not parts of the explanation but rather because, in giving explanations, we usually only state the relevant or interesting bits of the explanation, leaving the obvious parts implicit or unsaid
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
7. Board Game: I Spy In Common [Average Rating:5.75 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Counterexamples

Some arguments meet all of the positivists’ requirements to be an explanation but in fact are not explanations - these serve as counterexamples to the covering-law view
If these in fact fail to be explanations, it follows that the covering-law view is a false view of explanation

Common-Cause Cases

In some cases, we can derive a certain fact from certain other facts, not
because one explains the other but rather because they all have the same cause in common
 From the presence of certain spots and a law connecting spots with fever, we can deduce the presence of a fever
 But it’s a single illness that explains all of this, not what shows up in the argument

Covariation Cases

Common-cause cases are just one kind of covariation case
When two things occur together in a law-like manner, we can derive one from another, even though neither is the real explanation for the other
In many cases, one will be a mere symptom of the other or each will be a symptom of some third, underlying thing such as a common cause

 We can derive, using basic laws, the height of a flagpole from the length of its shadow
 But the length of the shadow isn’t part of the real explanation of the pole’s height

Salmon’s asymmetry problem is an instance of the sort of problem caused by covariation cases
Sometimes when A explains or causes B, one can derive A from B and vice versa
So the argument can start from either A or B and get us the other
But explanation does not work that way - it only starts with one of these

To do on your own: Covariation cases pose a problem for the covering law account of explanation. Come up with your own example of a covariation case (not from class or the readings) and explain why it’s a problem for the covering law view.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
8. Board Game: Pillz [Average Rating:6.33 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Overdetermination Cases

Sometimes there are present in some cases more than one thing which, on its own, would be sufficient to explain the explanandum
We can use either one to derive the explanandum
But only one of them is the real explanation

 Being male and taking the pill can each explain lack of pregnancy
 But in the case of a man taking the pill, it is being male that explains it, not his taking the pill

Criticism of the Symmetry Thesis

Hempel’s definition of explanation is not the only part of his account that seems to be in trouble from these cases
Symptoms of some underlying explanation, for instance, may be used to predict that underlying thing, but - as we’ve just seen - they may not be used to explain it
So the symmetry thesis is in trouble

Hempel’s Reaction to All of These Problems

Like in many other cases, Hempel anticipated many of these problems
And like in many other cases, his response was to basically accept the implausible consequences of his view rather than reject it
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
9. Board Game: True or False [Average Rating:4.18 Overall Rank:15198]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Explanation and False Laws

So far, we’ve seen plausible cases where the D-N view counts things as explanations which should not be
But does it fail to count some things as explanations which should in fact be?
Some philosophers of science would say "yes"

Note first of all that true, exceptionless general laws are rare outside of the natural sciences
Psychology, economics, and the social sciences seem to feature few, if any, such laws
Even in the physical sciences, though, things do not fit the D-N model so well
Many laws in, for instance, physics are in fact idealizations - they accurately describe an ideal situation but are strictly speaking false of the real world

Scientific theories in general feature a tradeoff between idealization and inexactness
The more exact a theory is, the more likely it will feature a number of idealizations and be accurate in some ways, inaccurate in others
The less exact it is, the more accurate it can be

Nancy Cartwright (1944- )
In How the Laws of Physics Lie, she argues that the fundamental laws of physics are useful falsehoods

Even if this is correct, these strictly-speaking false laws or theories seem to show up in our explanations of various phenomena
Yet we still seem to count them as explanations
If so, this looks like not only can scientific laws in fact be false, but parts of good explanations can be false as well!
This is most obviously seen when scientists still use a "rejected" paradigm instead of its successor in explaining phenomena
Scientists, for instance, still use Newtonian physics as opposed to relativistic physics to explain things
In fact, for some phenomena, scientists use Newtonian physics almost exclusively!

This is all rather strange, but we’ll look at a notion of theories later which will make much better sense of these ideas
It will then be clearer how perhaps something strictly false can be involved in explaining something that is true
Before we turn to theories, we will first examine various other views of explanation
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
10. Board Game: Cause & Effect [Average Rating:0.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Causal Accounts

With the demise of the Covering-Law account, causal accounts of explanation came back into vogue

Wesley Salmon (1925-2001)

Salmon noted that there is a temporal asymmetry in explanation - things earlier in time explain things later in time, but not vice versa
This points to causation as the key to explanation rather than argumentation
Causation is also asymmetrical in time, with earlier things causing later but not vice versa
Arguments are not asymmetrical in this way

To explain something on a causal account, roughly, is to give its cause
Like Hempel noted, though, this can be elliptical
Usually we are only interested in one or two causal factors that helped bring about a certain phenomena and leave all the other many factors unsaid

The Principle of the Common Cause: When there are events of a certain type occurring near-simultaneously at different places more frequently than should be expected if they occur independently of each other, then we should explain the occurrence of each in terms of a common cause
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
11. Board Game: Serenity: The Unofficial Board Game [Average Rating:6.40 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Unification Accounts

Although the covering-law view was the official view of the logical positivists, they also held to an unofficial view of explanation that they perhaps didn’t distinguish carefully enough from their own covering-law account
This is a view of explanation as unifying diverse phenomena under a few concepts, principles, entities, etc.

Philip Kitcher (1947- )

A view of explanation should:
 Show how it advances our understanding
 Enable us to comprehend and arbitrate disputes in past and present science (by judging the explanatory adequacy of beginning theories)

The covering-law view fails on both counts
On its own, deriving a conclusion from a set of premises does not necessarily advance our understanding of that conclusion
The covering-law view does not have enough resources to reliably distinguish better explanations from worse ones
So the covering-law view fails on both accounts

In addition, Kitcher shares Hume’s skepticisms about the notion of causation
So Kitcher, initially at least, rejects both covering-law and causation views
Instead, he opts for a version of the positivists’ unofficial view
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
12. Board Game: Risk: Balance of Power [Average Rating:5.89 Overall Rank:6426]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Unification Accounts

Kitcher: Explanation is simply connecting diverse facts by subsuming them under a set of basic patterns and principles
An argument explains something when it is part of the set of arguments which best unifies all the things we currently believe
This set is the explanatory store

Kitcher’s account, like the covering-law view, connects explanation with argumentation
Because it is more selective as to which arguments are explanatory, though, it may be able to avoid the problems suffered by the covering-law view

The Asymmetry Problem

Arguments going from the length of an object’s shadow to the height of the
object will not feature as part of the explanatory store
These arguments do not unify things enough - it is better to have principles that operate for the height of all flagpoles, not just ones that happen to be casting a shadow

Comparing Explanatory Power

Kitcher’s view can also deal in a similar way with explaining why accidental generalizations are not explanatory
In general, it can help decide between theories by comparing their explanatory power
This is done simply by comparing how unifying the explanatory stores of each theory are
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
13. Board Game: Choices [Average Rating:0.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Do We Have to Choose?

There is a tendency in philosophy of science now towards admitting a number of different kinds of explanation
Causal explanation and unifying explanation may be two kinds of explanation
If that’s true, the causal and unifying accounts aren’t really rivals, but are about different kinds of explanation
Salmon and Kitcher themselves now seem to accept both causal and unifying explanations as real kinds of explanation
In fact, we might even allow some explanations to be of the sort governed by the covering-law view

Admitting different sorts of explanation goes all the way back to Aristotle, so it shouldn’t be too surprising
 Efficient, teleological, formal, and material "causes"

Contextualist Views

On a contextualist view of explanation, a true statement is an explanation when it answers a question, but only in certain contexts
Sometimes, a causal statement might be an explanation, other times a unification or covering-law account might be what’s needed - it all depends on the question and the context

 Not just any question will do
 Answering "yes or no" questions does not give us any explanations
 Neither, perhaps, does answering a question for which you already have a plausible suggested answer
What about context? Are there any constraints there?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
14. Board Game: Context [Average Rating:3.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Van Fraassen’s Contextualism

Bas van Fraassen (1941- )

Van Fraassen adopts a kind of contextualist view
Normally, explanatory power is taken to be a reason to believe a theory
Science, people tend to think, is after explanations

Van Fraassen: This is all wrong
Science itself does not aim at explanation
Explanation is a matter of using science for your own purposes, not part of science itself
It is an application of science to satisfy your desires for certain descriptions

How good an explanation is is simply a matter of how good an answer it is to your question
How good an answer it is is relative to the context and always in comparison with a contextually-specified range of alternatives
A really good explanation counts as really good because it is really good in comparison with a set of specified alternative explanations

Potential Problems for van Fraassen

Is it really plausible, though, that science does not aim at explanation at all?
Don’t scientists sometimes choose new theories precisely because they have more explanatory power than their predecessors?
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
15. Board Game: Spot the difference lotto [Average Rating:3.67 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Kuhn’s Contextualism

According to Kuhn, each paradigm or theory has its own standards for explanation
(Kuhn thinks each paradigm also has different standards as to what causation is, but this is perhaps less plausible)
Scientists therefore look for different things in a hypothesis depending on their paradigm or theory

When working with one theory or paradigm, what we are looking for in an
explanation will be different from someone working in another
Sometimes we want a cause, sometimes something else
It all depends on the theory or paradigm

This doesn’t mean that it is entirely arbitrary what we are looking for when we look for an explanation
There may be good reasons for the differences between theories or paradigms rooted in the theories or paradigms themselves
 Ex: Differences in what supposedly exists, the subject matter, etc.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
16. Board Game: Cooking Up Sentences [Average Rating:4.50 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Are All Theories Sentences?

So far, many of the accounts we have seen of explanation seem to assume that a theory is something like a sentence or set of sentences
If treated as sentences, it looks like many of our scientific theories may well be too exact to be strictly true and yet may still play a role in explanation

Meaningful, declarative sentences are in general either true or false, period
It’s hard to measure the accuracy of a sentence when it is false
If false, the sentence has in some way missed its mark
There are, however, non-linguistic sorts of representation that work slightly differently
Not all representations are sentences or pieces of language
Maps, for instance, admit of degrees of accuracy in various respects
 Maps are not simply true or false
 A map may fail to be 100% accurate in every respect yet still meet its goal as a map

Many ideas in science are not expressed entirely in linguistic form but rather by using models

Ronald Giere (1938-)

Advocates a view of theories (and science generally) based on this notion of a model. (Also advocated by philosopher of science and physics Paul Teller, my now-former advisor )
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
17. Board Game: Super Model [Average Rating:0.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Models

A model is, roughly, a structure which is intended to represent a target in certain respects by being similar to it in certain ways and to a certain degree

Note that, unlike sentences, a model’s accuracy is a matter of having a kind of similarity relationship with the target system
What kind depends on the model, its target, and how we are supposed to use it
Sentences themselves can be used to express a model

What matters for the accuracy of a model is that the degree of similarity between the model and real world target is close enough for the purposes at hand
So whether a model is accurate enough or not depends on how we are going to be applying it to the real world

Other considerations that might be taken into account in evaluating a model:
 Simplicity
 Ease of use or understanding
 Explanatory power
 Personal, social, or political considerations
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
18. Board Game: Jim Prentice Electric Baseball Model 505 [Average Rating:6.00 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Models

The same domain of facts can be described using different, apparently inconsistent models
 Water as a continuous, incompressible fluid to model water flow
 Water as a collection of particles to model heat

Different models can be used to represent different aspects of the same system
We can even have models of models!

Known inaccuracies in a model consist of those respects in which the model is an approximation or idealization
 Example of idealization: Using Newton’s law of gravitation to determine the speed of a falling rock at a point when it has been dropped only a short distance, treating it as if it was falling in a vacuum, etc.

 Other examples: Boyle’s law, models treating objects as point-particles, etc.

Distortions here make description and calculation easier and don’t significantly affect our accuracy in describing the facts of interest about the real world
A large part of scientific apprenticeship involves learning the craft of model-building

Kinds of Models

Kinds of models include
 Physical models
 Mathematical models
 Computer models

Physical models include models of molecules, model
planes, etc.
Mathematical models are abstract mathematical structures supposed to represent key features of real systems in the world
These can include equations and other pieces of mathematics
Computer models might be more or less like a physical or mathematical model
Sim games can be seen as kinds of interactive models
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
19. Board Game: Experiment Expansion [Average Rating:6.39 Unranked] [Average Rating:6.39 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Laws and Models

Many people have not been sure whether to count, say, Newton’s fundamental laws as definitions or empirical hypotheses about the world
Are they true by definition or could they turn out to be false?
Giere: The real situation is a little more complicated than this

Many models are characterized by certain laws or defined in terms of them
In these models, such laws are true by definition - but only in the model, not the real world
Laws are used to construct models which can then be applied to the real world
Laws, then, get applied to the real world only indirectly through a model, but the model itself can still be accurate enough in the proper respects
So to say with Cartwright that laws are useful lies is a little misleading

Experiments as Models

Experiments can also be seen as a kind of physical realization of various models
In an experiment, we bring the world as close as possible to the model
We need to do this in order for our scientific laws to apply since they only directly apply to the situation of the model, not the real world in general

An experimental situation itself, then, can act as a kind of model for real world happenings
 Example: Making a model of an airplane and putting it in a wind tunnel to model the interactions of a real airplane with real air currents
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
20. RPG: D6 Adventure [Average Rating:7.08 Overall Rank:465]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Theoretical Hypotheses

Theoretical hypotheses are sentences that state the ways in which a model may be said to be similar to a body of phenomena in certain respects and to what degree of accuracy
Theoretical hypotheses, then, tell us about the accuracy of a model
Even though a model may not be true or false, theoretical hypotheses can be

Theories and Models

Should theories be identified with descriptions of models, models themselves, theoretical hypotheses, or some combination of these?
Godfrey-Smith: Some theories are linguistic, others are models

Giere: If we want to follow the way scientists actually use the word "theory", we need to think of it as involving both aspects
A theory is a collection of models or model-building toolkits, plus theoretical hypotheses describing how these apply to the real world
Laws and other bits of theoretical structure can be used as tools to construct models
Models themselves can be used to construct even further models
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
21. RPG Item: The Adventurer's Toolkit [Average Rating:7.52 Overall Rank:389]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
Theories and Paradigms

A new theory provides a new set of tools for capturing phenomena in some way better than our old tools
Paradigms (or research programs or traditions) can be seen as fundamental, large-scale model-building toolkits
Extending a paradigm can then be seen as a matter of applying the tools of the paradigm to build new models or refine old ones

Paradigms in the narrow sense can be seen as models of science for further models to follow
Science, on this sort of view, is essentially a process of model-building and -refining

We now have some of the tools to understand why some old, "rejected" paradigms such as Newtonian physics hang on despite being apparently "replaced" by newer paradigms
The old theory or paradigm, which is still being used, has some of the right structure to capture in a significant way some aspect of the real world
So it is kept as a model for certain purposes
Some argue this shows that there is no clear distinction between paradigms

Theories and Explanation

A theory, then, can feature in an explanation of a certain phenomena when
one of its models (or a model constructed with it) is close enough for our purposes in the relevant way to the real structure in the world which explains the phenomena
This is why laws which might be thought of as strictly false can appear in explanations for things that are true
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
22. Board Game: Science Fusion, the Elements of the Scienauts [Average Rating:3.45 Unranked]
David Spencer
United States
San Ramon
California
flag msg tools
MORE UberText!!!
badge
UberText!!!
Avatar
mbmbmbmbmb
A Model of Science

Giere’s view of science, of course, is supposed to apply to itself
Science as a model-building enterprise is itself a model and hence will be accurate in certain respects and not so much in others
It doesn’t seem to deal well with some scientific classification systems, for instance
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
Loading... | Locked Hide Show Unlock Lock Comment     View Previous {{limitCount(numprevitems_calculated,commentParams.showcount)}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
{{error.message}}
{{comment.error.message}}
    View More Comments {{limitCount(numnextitems_calculated,commentParams.showcount)}} / {{numnextitems_calculated}} 1 « Pg. {{commentParams.pageid}} » {{data.config.endpage}}
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.