Introductory textbooks in biology, physiology and statistics present material to the student without origins or qualifications. That is, they present autonomous facts, creating the reality of science. Psychology textbooks do not do this when they present psychology. Instead, they give evidence and they qualify and hedge the claims made. Psychology texts write autonomous facts of biology and statistics, just as biology and statistics texts do, but move away from autonomous fact-writing when material is recognized as psychological. The boundaries between disciplines are created in the modalities of the statements in the textbooks. The `fact' status of material on the activities of neurons in colour vision and feature detectors changes when this material is written as physiology in physiology textbooks and when it is written as psychology in psychology textbooks. Statistics, biology and physiology are presented as `certainty sciences' both in the home discipline textbooks and in psychology textbooks, but acceptance of material as psychological leads to the presentation of uncertainty. `Certainty writing' in psychology is not simply a reflection of a users' `certainty trough', but of textbook writing practices within the certainty sciences themselves. The failure of psychology textbooks to present autonomous facts of psychology is not an indication that no well fabricated knowledge exists in psychology, but rather that psychology habits of textbook writing involve both relevance and evidence. Maintaining relevance to everyday experience but distinguishing the claims of science from the claims of the everyday produces a form of textbook writing which does not present certainty, but which presents evidence instead.