THIS PAPER PRESENTS THE SUCCESSIVE STAGES USED IN ONE SPECIFIC ATTEMPT TO APPLY ASPECTS OF SYSTEMS THEORY TO FORMULATE A PRELIMINARY THEORY ACCOUNTING FOR THE PRESENCE OF VIOLENCE AS A CONTINUING ELEMENT IN THE SOCIAL STRUCTURE OF THE NUCLEAR FAMILY. THE SEQUENCE STARTS WITH A ROUGH BLOCK DIAGRAM TO PROVIDE AN INVENTORY OF POSSIBLY RELEVANT VARIABLES AND TO CRUDELY SUGGEST THEIR INTER-RELATIONSHIPS AND SOMETHING OF THE CAUSAL FLOW. THE NEXT STEP CONSISTS OF ATTEMPTING TO PULL TOGETHER RELATIONSHIPS AND ASSUMPTIONS IN THE FORM OF A SET OF INTER-LINKED PROPOSITIONS WHICH, TOGETHER, SEEM TO ACCOUNT FOR THE STABILIZATION OF VIOLENCE IN THE FAMILY SYSTEM. PROCESSES OF LABELING, SECONDARY CONFLICT, REINFORCEMENT, SELF-CONCEPT FORMATION, AND ROLE EXPECTATIONS ARE IDENTIFIED AS KEY ASPECTS OF THIS PROCESS. FINALLY, A FLOW CHART IS DEVISED TO MAKE EXPLICIT THE NATURE OF THE BRANCHING AND FEEDBACK PROCESSES WHICH PROVIDE THE DYNAMIC ELEMENTS OF THE SYSTEM. IN THE COURSE OF THE FLOW CHART ANALYSIS, THE THEORY IS AMPLIFIED BY THE INCLUSION OF MORPHOGENIC PROCESSES. THE THEORY REPRESENTED BY THE FINAL FLOW CHART IS SUPERIOR TO THE BLOCK DIAGRAM USED INITIALLY BECAUSE IT REPRESENTS THE DECISION PROCESSES IN RELATION TO SYSTEM GOALS AND THE LEARNING PROCESSES WHICH GIVE THE FAMILY THE POTENTIAL FOR BEING AN ADAPTIVE SYSTEM. A FINAL SECTION OF THE PAPER OUTLINES THE IMPLICATIONS OF A SYSTEMIC APPROACH FOR RESEARCH DESIGN AND RESEARCH TECHNOLOGY. (AUTHOR ABSTRACT MODIFIED)