Plants & Human Affairs (BIOL106)  -  Stephen G. Saupe, Ph.D.; Biology Department, College of St. Benedict/St. John's University, Collegeville, MN 56321; ssaupe@csbsju.edu; http://www.employees.csbsju.edu/ssaupe

The Nature of Science

I.  What is science?  
    Latin - scientia, meaning to know or have knowledge.  By this definition, everything is science!

A.  Comparison of disciplines

• Natural science - scientific study of phenomena that would exist in the absence of humans (i.e., botany, chemistry, biology, geology, physics, astronomy, mathematics)

• Social science - scientific study of phenomena arising from the activity of humans (i.e., government, sociology, economics); 

• Humanities - studies of human activities/achievements (i.e., art, English, theater, theology)

Which is best?  None...each has its particular purpose and should not be used for the others. But, what is science?

B.  What science is.  Among the many possible definitions, Science�.  (don't memorize these)

• tells us how many and how big, but not how beautiful

• is the systematic gathering of information from observation or experiment

• is an exploratory activity for the purpose of understanding the natural world (P. Medawar)

• is organized skepticism (M. Curd)

• is a self-corrective discipline

• is scrupulously applied common sense (Huxley)

• is the search for understanding, which is achieved by statements of natural laws applicable to a wide variety of situations, and that are tested experimentally (Goldstein & Goldstein)

• seeks to organize knowledge in a systematic way, endeavoring to discover patterns of relationships among phenomena and processes

• strives to provide explanations for the occurrence of events

• proposes explanatory hypotheses that must be testable, that is, accessible to the possibility of rejection

• is an interconnected series of concepts and conceptual schemes arising from experiment and observation and yielding new observations and experiments (J. Conant)  

C.  Stating Hypotheses
    Typically there may be more than one hypothesis to explain a particular event or observation.  These are listed H₀...H_n.  For example, there are multiple explanations, though not all are equally likely, to explain what happened to Tom Jones.  These include:

H1 - Tom swam perpendicular out of the rip tide;
H2 - Tom was saved by a dolphin;
H3 - Tom had a surge of adrenalin and was able to swim against the current;
H4 - Tom was saved by a rogue wave that pushed him to shore; and so on.

What scientists essentially do is to disprove or eliminate all the alternative hypotheses leaving the one "correct" explanation.  Though in practice this sounds easy, it can be difficult to know when all potential hypotheses have been considered.

D.  The Nature of Proof
    You can never �prove� anything in science.  The reason for this is that impossible to test a given hypothesis under all possible conditions and circumstances.  Thus, we can never be absolutely certain that a hypothesis is True.   Consider the following "truth table":

If a hypothesis is true, then when we test the hypothesis, any predictions that are generated from the predication must also be true.  For example, let's consider the hypothesis "all green apples are sour."  If this hypothesis is true then we predict that every green apple we eat will be true.  Thus, the predictions from a true hypothesis will also be true.  However, if the hypothesis is false (which clearly it is), when we may still have true predictions (e.g., we eat a thousand green apples they might all taste sour) until we sample our first sweet "Granny Smith" apple.  Thus, a false hypothesis can give rise to both false AND true predictions.  What this means is that we can never prove a hypothesis to be true because true predications can be generated from both false and true hypotheses. 

Conversely, we CAN prove an idea to be wrong since it only takes one case to show that a hypothesis is not true.   To take advantage of the ability to disprove hypotheses, we often state them as a null hypothesis, H0, that is, we state that there is no difference between what we expected/hypothesized and what we found.  And, if we prove this to be false, then that means there must (�proved�) be a difference between the two.  This is a commonly used technique by �statisticians".  The take-home-lesson from this is that we can support an hypothesis, but we can't prove it. Don't say that you�ve �proved� your hypothesis �  instead, say that you've �supported� it.

VI.  Steps of the "Scientific Method"

1. Make observations/gather information;
    

2. Develop a hypothesis.
    

3. Generate predictions from the hypothesis that will be tested with an experiment
        Predictions are generated by deductive logic - formulated as "If...then..." statements.  In deductive logic, we start with a generalization and make a specific conclusion.  Inductive logic is just the opposite, you start with a few specific observations and make a generalization;
    

4. Design an experiment to test the hypothesis.  Don't forget a control group;
    

5. Collect data (results) 
    

6. Make conclusions;
    

7. Report results;
    

8. Repeat ad infinitum modifying hypotheses as necessary and/or generating new ones.

VII. Analyzing Experiments:  Some considerations

A.  The experiment must be repeatable
    To confirm conclusion and guard against one-time events (e.g., many examples in Secret Life of Plants).

B.  Avoid Bias in interpreting the Results
    Scientists are humans and subject to biases and prejudices.  Can minimize by doing double blind experiments and including placebos.  There are two major types of bias:

1. Unintentional - unconscious slanting of data to get result (e.g., Pavlov assistant, Morton & cranial capacity)

2. Intentional - conscious manipulation of data = dishonesty!  Ranges from "prettying" up data (ignoring anomalous results) to outright fabrication.  Why would a scientist cheat?    Reasons:  competition for grant money; pressure to publish; glory.  Fraud is especially destructive in science since there is no time to check everyone's work, must assume it is correct.  It's hard to imagine why a scientist would cheat since the self-corrective nature of science will eventually expose the cheater.

VIII.  Biology has few laws
    The more data that support a hypothesis, the more strength that it has, and we presume the greater "truth".  The general scheme of increasing strength of support for an idea is: hypothesis → theory  → law.  In practice, there is no clear distinction between law, theory and hypothesis.  For example, even though you hear of the �theory� of evolution, it doesn�t mean that biologists consider it is any less a fact that say, the �law� of gravity.

    Thus, biology is a statistical science.  Generalizations in biology are probabilistic - because there are often exceptions.  Why?  because living organisms are so diverse and exhibit exceptional variety.             


VIII.  A Case Study:  Ignaz Phillip Semmelweis and Childbed Fever (for more information see the article by Brown and Williams, 1990)   (not directly covered on exam)

1. Background & Problem: 
       Semmelweis (1818-1865) was a Hungarian physician; received a medical degree and midwifery from the University of Vienna (1844).  He worked at Vienna General Hospital Obstetrical Clinic.  He noticed that many women in the clinics died of childbed fever.  The cause was unknown but was "hypothesized" to be caused by miasma.  Because of the high risk of dying from this fever which was associated with hospitals, most women had births at home.  Clinics were just for poor, unwed mothers.  Childbed fever is now known to be caused by Streptococcus pyogenes (Staphylococcus) - like one that killed Jim Henson (Mr Muppet) and the face-eating bacteria.
   

2. Observations:
       There were two clinics at the hospital � one with medical students and physicians and the other with midwives.  The mortality rate was high in the First clinic that was tended by physicians and students.  The mortality was low in the Second clinic staffed by midwives.  Semmelweis observed that in the first clinic the physicians and students would perform autopsies, exams, and examine sick patients.  The midwifes cared for patients but performed no autopsies.  The mortality in the First clinic was high (about 10%) but low in the Second (about 2%).  The First clinic admitted patients on Saturday, Sunday, Tuesday and Thursday, while the midwife (Second) clinic admitted patients on Monday, Wednesday and Friday.  The delivery practices in the two clinics were the same and the deaths occurred in patients admitted on Saturday, Sunday, Tuesday and Thursday.  One of Semmelweis teachers died from infection received during autopsy; symptoms similar to childbed fever.
   

3. Hypothesis:  
       The medical students and teachers are passing the disease from patient to patient
   

4. Alternate hypotheses:  
       The patients sick before arrival.
       Miasma caused disease.   
   

5. Predictions:  
       IF the medical students and teachers are passing the disease from patient to patient THEN, washing hands should prevent disease
   

6. Experiment: 
       Semmelweis ordered everyone to wash hands with chloride of lime
   

7. Result:  
       Death rate dropped.  
   

8. Conclusion/Evaluation:  
   Semmelweis - Problem solved
   Others - Coincidence (especially because they stopped washing hands and death rate rose again.  End result:  Semmelweis' work NOT accepted especially because he didn't publish his work.  He spoke in 1850 at Med. Society of Vienna and physicians were on the verge of acceptance, but adidn't because he hadn't published (at least until much later in a rambling incoherent way).  He eventually died in an insane asylum at 47.  
   
            For a dialog about Semmelweis, click here

     I.  Take-Home-Lessons:

1. Societal acceptance of/impact on science

2. Development of hypothesis; tested by gathering evidence

3. Control vs. experimental groups - indirectly

4. Importance of publication

5. Standard scientific method not always - usually have an opinion about what the endproduct of experiment is...do experiment to support/prove yourself wrong.

6. The time was not ripe for understanding of his discoveries

IX.  Case Study - Frog and Toad in the Garden
    This wonderful children's story by Arnold Lobel provides a look at experimental design and inappropriate conclusions (non-causal correlations) based upon a poorly designed (no control group) experiment.
            click here for details
 

X.  Case Study � Card Playing (Science is Predictive) 
    We will see if you can determine the pattern that I arrange in a deck of playing cards.  This exercise should focus on concepts of falsifiability, certainty, proof, alternate hypotheses, laws and biology.              
       click here for a mini-exercise
 

XI.  Case Study - Dodos and Trees
    This story is included because it demonstrates nicely how scientists can study past events. It is also a good example of non-causal correlations.  

A.  Observations.
    On the island of Mauritius, the native home of the now extinct dodo, Calvaria trees have not germinated naturally for about 300 years (the age of the youngest tree), which is about the same time as the dodo went extinct.

B.  Question.
    Was the extinction of the dodo responsible for the failure of the Calvaria tree seeds to germinate?

C.  Hypothesis.
    Germination of the seeds of the Calvaria tree only occurred naturally after they passed through the gut of a dodo.

D.  Collect evidence/experiments.
    We haven�t mentioned it yet, but technically what scientists do to support or disprove a hypothesis is to determine if predictions that arise from their hypothesis are true.  Let�s list some predictions from the hypothesis.  IF the dodo is responsible for the failure of the Calvaria seeds to germinate, THEN......

1. there should be no trees younger than 300 years when the dodo went extinct.  (This is true)

2. the dodo should have been big enough to eat the seeds (True)

3. the dodo was small enough to prevent completely crushing the seeds (True, demonstrated by inference to other bird studies);

4. dodos would eat seeds (True - seeds found in fossils)

5. passage through a dodo gut should cause the seeds to germinate (True - from studies force-feeding turkeys).

E.  Conclusion.
    The author concluded that the extinction of the dodo has lead to the near extinction of Calvaria trees because the seeds have a very hard pit that can not be broken down.  The evidence seems to be very strong in support of the hypothesis. 

    But, does this prove our hypothesis?  No, for at least two reasons: 

1. Remember we can never �PROVE� our hypothesis - it�s always with the realm of possibility that an alternative is true - for example, perhaps the seeds aren't germinating because of a climate change, or a new disease, or pollution in the water, or deforestation, or introduced species, etc.  In fact, this article, which was published in one of the premier journals of American science, stimulated a flurry of research and now has been discredited.  
    

2. The experiment was poorly designed/executed [small sample size, inadequate controls, inadequate background information (small trees ARE known), inappropriate graph interpolation]

For more, check out David Hershey's essay, "The Widespread Misconception that the Tambalacoque or Calvaria Tree Absolutely Required the Dodo Bird for its Seeds to Germinate" in Plant Science Bulletin 50: 105 (2004).

Dodo Article Analysis - click here

XII.  Case Study - van Helmont and his willow. (not directly covered on exam)
    We will analyze this classic experiment that was very important in the history of plant physiology.  Interestingly, it had some �flaws�. 
                 Click here for details.

    
XIII.  Case Study - Willow Cuttings   (not directly covered on exam)
                 Click here for details.
 

References:

• US Dept of Health and Human Services.  1992.  Curiosity is the Key to Discovery.  The Story of How Nobel Laureates Entered the World of Science.  

• Brown, Jr. WE & RP Williams.  1990.  Ignaz Semmelweis & the importance of washing your hands.  American Biology Teacher 52:  291 � 293.  

Last updated:  10/06/2008 / � Copyright  by SG Saupe