Welcome to Gaia! :: View User's Journal | Gaia Journals

Given: The best way to do research on a phenomenon is to study it in a context as close as possible to that in which it was discovered, for best relevance.

Given: Studying a phenomenon in the context in which it was discovered is often very difficult, because the phenomenon itself may be very noisy, and the environment almost certainly will be noisy also. So most phenomena are studied at one or another degrees of abstraction away from the original context.

Given: every level of abstraction away from the original context takes away a certain quantum of validity.

Given: Where possible, computational models should be used to study these phenomena, because they are capable of processing enormous amounts of material at relatively high speed. (Sure, it might take half an hour to simulate a chemical cascade reaction that takes microseconds, but that's still pretty amazing.)

Given: The computational models we currently have are vastly incomplete.

Given: Since our computational models are incomplete, in the life sciences, the next level of abstraction closer to the original context is in vitro studies. These are done with tissue samples, which are usually taken from living animals in terminal procedures (i.e., the animal is killed in the process), usually under a lethal dose of anaesthesia.

Given: In vitro studies, while more realistic than in silico (that is, computerized) studies, are still isolated from the complexities of life inside a whole organism. Even the most sophisticated techniques cannot capture the whole of it.

Given: The next level of abstraction is in vivo study in an animal model. This almost always involves invasive procedures, or at least procedures requiring foreign substances to be applied to the exterior of a living animal.

Given: Ideally, life-sciences studies of humans would be done through studies of humans.

Given: Humans are difficult to control, and invasive research on humans is almost universally prohibited by sworn codes of conduct, professional regulations and national and international law. Exceptions are extremely rare and fraught with confounding variables.

Now: my conclusion is, therefore, that if we want to do biomedical research that's relevant to human beings, in any study in which our computational models do not square very precisely and accurately with field data, we must do some kind of invasive technique - tissue sampling and culturing, or biopsies, or implants of electrodes.

Now: there are some givens relating to research using living animals. For example:

Given: Any invasive procedure conducted on an animal, even under anaesthesia, causes a certain disturbance of physiology.

Given: Any distress or suffering experienced by an animal causes further disturbance of physiology.

Given: Any disturbance of physiology in an animal model of normal physiology will disturb the validity of the study.

However, it is my view that these costs are outweighed by the benefit of using animal models rather than human models.

If you feel that the benefits derived from the biomedical study of animal models do not outweigh the costs of using animal models, then I request that you reply to this journal with a suggestion of how equivalent, scientifically valid research could be conducted in a manner that provides similar benefits but does not violate sworn codes of conduct, professional ethics, or national or international law.