It’s the scientific method. You have to know what you are testing. The null hypothesis is that the relationship you think is there isn’t there. You need the alternative hypothesis to describe the thing you are looking for. You need statistically significant evidence to support the alternative hypothesis or else you can not disprove the null hypothesis.
If you don’t clearly and accurately define the null and alternative hypotheses, at the end of the experiment, you have no idea what you’ve just concluded.
@wundayatta I suggest that we reject or refuse to reject the null hypothesis based on the experimental results and the statistical analysis of the data.