Summary Scientific values
Text 1: Descriptive and normative claims in science and engineering
Imperatives: instruct to make something the case, end with !
Questions: inquire whether something is the case, end with ?
Declaratives: declare something to be the case, end with .
There are 2 types of declarative claims:
1. Descriptive claims: claims that potentially describe states of affaires, phenomena or facts.
They are either true or false which can be shown by test or experiment. Mostly objective.
2. Normative claims: claims that give a recommendation. You can either agree or disagree.
• Explicitly: prescriptive claim. Explicitly recommends an action, usually contains
should.
• Implicitly: evaluative claim. These have the form of value judgements
Normative-descriptive and subjective-objective are independent. Apart from the salient descriptive-
objective and normative-subjective, there are also examples of:
• Descriptive-subjective: Green to me is closer to blue than to yellow. This claim is subjective in
the sense that its truth/falsity lies entirely ‘within you’. By experiment a true/false can be
determined.
• Normative-objective: One should not kill. Normative-objective statements are normatively
valid for everyone. It depends on your views whether this is true. Almost all moral
judgements fall into this category.
A stipulative definition is when a word is introduced into a language, with an explication of what it
stands for/how it is supposed to be used. They can be seen as prescriptions.
That one cannot validly derive a normative conclusion from descriptive arguments is called the ‘is-
ought gap’, where is refers to descriptive and ought to the normative, > dogma?
,Text 2 – The validation of scientific claims
Valid argument = argument where conclusions follow from premises. Truth goes in, truth comes out.
Sound argument = valid argument where premises are true. The truth depends on 2 aspects:
1. Validity of the argument is a matter of its form (it is independent of what the propositions
occurring in it say).
2. Truth of the premises is a matter of their content (it depends on what the specific
propositions say).
Enumerative induction:
This is false, this only goes for metal so far > therefore it is a generalization.
An asymmetric conclusion is one that cannot be turned around, a symmetric conclusion can be.
A test for a condition where 2 or more theories predict different things to happen is called a crucial
experiment. By falsification of 1 theory, one could verify the other. However, this is still invalid
verification because it is impossible to predict all possible theories and falsify them all.
The hypothetico-deductive method. Universal statements are proposed as hypotheses (truth value
is not yet certain). Next, a prediction of a phenomenon is made from generating a hypothesis. Finally,
the prediction is compared with empirical facts to draw a conclusions about the statement. So:
If hypothesis H is true, then if condition C obtains, we should observe empirical fact E.
Verification with the hypothetico-deductive method is invalid. However falsification is valid.
Sophisticated falsification: for any theory, multiple assumptions are required to predict what will be
the case in a particular empirical condition. If the prediction is not observed, we can conclude that
either the core theory or one of the assumptions is false.
, Deductive reasoning (deductively valid)
• The truth of the premises guarantees the truth of the conclusion
• All information in the conclusion is already contained in the
premises
• Adding premises cannot change the conclusion > monotonic
Inductive reasoning (ampliative)
• The truth of the premises supports the truth of the conclusion
• The conclusion contains information that goes beyond the premises
• Adding premises can change the conclusion > non-monotonic
Deductive reasoning is not fit to model reasoning with respect to scientific knowledge because there
is no proof. Every theory at some point can be proven to be false, scientific knowledge is fallible and
tentative. Reasoning in science is mostly inductive or ampliative.
1. Observation: gathering empirical facts from the world
2. Induction: proposing laws, theories and mechanisms to derive a hypothesis
3. Deduction: derivation of consequences, predictions are made concerning other facts we
expect to observe in particular circumstances if the hypothesis is right
4. Testing: creating circumstances to record what is actually observed
5. Evaluation: compare the observations with predictions to judge the truth of the hypotheses.
The 3 reasoning phases of the empirical cycle are characterized by distinct reasoning forms.
• The reasoning in the prediction phase is deductive.
• The evaluation phase however cannot be deductive. Instead it will have to be inductive in
the broad sense
The context of justification part of the cycle contains the prediction and evaluation phases
The reasoning underlying the hypothesis formation is abduction
Text 1: Descriptive and normative claims in science and engineering
Imperatives: instruct to make something the case, end with !
Questions: inquire whether something is the case, end with ?
Declaratives: declare something to be the case, end with .
There are 2 types of declarative claims:
1. Descriptive claims: claims that potentially describe states of affaires, phenomena or facts.
They are either true or false which can be shown by test or experiment. Mostly objective.
2. Normative claims: claims that give a recommendation. You can either agree or disagree.
• Explicitly: prescriptive claim. Explicitly recommends an action, usually contains
should.
• Implicitly: evaluative claim. These have the form of value judgements
Normative-descriptive and subjective-objective are independent. Apart from the salient descriptive-
objective and normative-subjective, there are also examples of:
• Descriptive-subjective: Green to me is closer to blue than to yellow. This claim is subjective in
the sense that its truth/falsity lies entirely ‘within you’. By experiment a true/false can be
determined.
• Normative-objective: One should not kill. Normative-objective statements are normatively
valid for everyone. It depends on your views whether this is true. Almost all moral
judgements fall into this category.
A stipulative definition is when a word is introduced into a language, with an explication of what it
stands for/how it is supposed to be used. They can be seen as prescriptions.
That one cannot validly derive a normative conclusion from descriptive arguments is called the ‘is-
ought gap’, where is refers to descriptive and ought to the normative, > dogma?
,Text 2 – The validation of scientific claims
Valid argument = argument where conclusions follow from premises. Truth goes in, truth comes out.
Sound argument = valid argument where premises are true. The truth depends on 2 aspects:
1. Validity of the argument is a matter of its form (it is independent of what the propositions
occurring in it say).
2. Truth of the premises is a matter of their content (it depends on what the specific
propositions say).
Enumerative induction:
This is false, this only goes for metal so far > therefore it is a generalization.
An asymmetric conclusion is one that cannot be turned around, a symmetric conclusion can be.
A test for a condition where 2 or more theories predict different things to happen is called a crucial
experiment. By falsification of 1 theory, one could verify the other. However, this is still invalid
verification because it is impossible to predict all possible theories and falsify them all.
The hypothetico-deductive method. Universal statements are proposed as hypotheses (truth value
is not yet certain). Next, a prediction of a phenomenon is made from generating a hypothesis. Finally,
the prediction is compared with empirical facts to draw a conclusions about the statement. So:
If hypothesis H is true, then if condition C obtains, we should observe empirical fact E.
Verification with the hypothetico-deductive method is invalid. However falsification is valid.
Sophisticated falsification: for any theory, multiple assumptions are required to predict what will be
the case in a particular empirical condition. If the prediction is not observed, we can conclude that
either the core theory or one of the assumptions is false.
, Deductive reasoning (deductively valid)
• The truth of the premises guarantees the truth of the conclusion
• All information in the conclusion is already contained in the
premises
• Adding premises cannot change the conclusion > monotonic
Inductive reasoning (ampliative)
• The truth of the premises supports the truth of the conclusion
• The conclusion contains information that goes beyond the premises
• Adding premises can change the conclusion > non-monotonic
Deductive reasoning is not fit to model reasoning with respect to scientific knowledge because there
is no proof. Every theory at some point can be proven to be false, scientific knowledge is fallible and
tentative. Reasoning in science is mostly inductive or ampliative.
1. Observation: gathering empirical facts from the world
2. Induction: proposing laws, theories and mechanisms to derive a hypothesis
3. Deduction: derivation of consequences, predictions are made concerning other facts we
expect to observe in particular circumstances if the hypothesis is right
4. Testing: creating circumstances to record what is actually observed
5. Evaluation: compare the observations with predictions to judge the truth of the hypotheses.
The 3 reasoning phases of the empirical cycle are characterized by distinct reasoning forms.
• The reasoning in the prediction phase is deductive.
• The evaluation phase however cannot be deductive. Instead it will have to be inductive in
the broad sense
The context of justification part of the cycle contains the prediction and evaluation phases
The reasoning underlying the hypothesis formation is abduction
