The official Predator Free New Zealand programme launched in 2016 is based on a hugely inspiring, aspirational ambition to eradicate all invasive rodents (rats Rattus norvegicus and R. rattus but not mice Mus musculus), mustelids (stoat Mustela erminea, ferret M. furo and weasel M. nivalis) and possums (Trichosurus vulpecula) from throughout New Zealand by 2050. Others had already been doing predator control for years, but this campaign has caught the public imagination as no previous operation ever has. It is achieving some impressive results at local scales, to well-deserved acclaim. But its underlying philosophical world view is less often discussed, which, I argue, poses a risk to its prospects of long-term, national-scale public support. World views matter much more than we usually recognise because they determine the questions we ask and the answers we consider reasonable. The history of environmental management in New Zealand offers some thought-provoking examples of programme managers unconsciously committed to unhelpful world views. Some overlook hidden assumptions, e.g. that top-down methods of imposing artificial mortality can exceed the high natural mortality of resilient pest species such as rats, stoats, rabbits (Oryctolagus cuniculus), or deer (Cervus elaphus). Some ask the wrong questions, such as how to find better ways to kill pests rather than how to reduce the numbers to be killed, which is usually controlled by food supplies from the bottom up. Some favour the wrong conclusion, such as when an observed change in pest numbers or distribution is attributed to suppression by artificial means even when the natural means are unknown. The philosophy of reasoning suggests that the PF2050 programme could best be considered as a game of two halves. First, short-term prevention of damage to native values by existing top-down suppression that cannot eradicate pest populations but can at least protect the most vulnerable native fauna until we can think of better means to save them. Second, longterm removal of pest populations by supplementing suppression with unknown future methods of minimising pest fertility and immigration.