Because people wish to preserve their health and do something equivale
nt for ecosystems, the metaphor of ecosystem health springs to mind. T
his paper presents the argument that it is a mistake for environmental
scientists to treat this metaphor as reality. First, the metaphor fai
ls because it misrepresents both ecology and health science. Ecosystem
s are not organisms, so they do not behave like organisms and do not h
ave properties of organisms such as health. Also, health is not an ope
rational concept for physicians or health risk assessors because they
must predict, diagnose, and treat specific states called diseases or i
njuries; they do not calculate indexes of health. Second, attempts to
operationally define ecosystem health result in the creation of indexe
s of heterogeneous variables. Such indexes have no meaning; they canno
t be predicted, so they are not applicable to most regulatory problems
; they have no diagnostic power; effects on one component are eclipsed
by responses of other components; and the reason for a high or low in
dex value is unknown. Their only virtue is that they reduce the comple
x array of ecosystem responses to various disturbances to one number w
ith a reassuring name. A better alternative is to assess the real arra
y of ecosystem responses so that causes can be diagnosed, future state
s can be predicted, and benefits of treatments can be compared.