Here's why:

* Not all experiments aim to establish cause and effect: Some experiments simply explore a phenomenon, test a new technique, or gather preliminary data. In these cases, a control might not be relevant.

* Control is not always feasible: In certain situations, creating a control group might be impossible or unethical. For example, studying the effects of a rare disease on human subjects.

* Some experiments are observational: Observational studies analyze pre-existing data without manipulation, making controls irrelevant.

However, controls are highly beneficial in many experiments:

* Establishing causality: Controls help determine if the observed effect is truly due to the experimental treatment or a confounding factor.

* Measuring baseline: Controls provide a reference point for comparison, allowing researchers to measure the magnitude of the treatment's effect.

* Increasing reliability: Controls help minimize the influence of extraneous variables, making the results more reliable and interpretable.

In summary, controls are a valuable tool in scientific experimentation, but they are not universally required. The need for controls depends on the specific research question, the nature of the experiment, and ethical considerations.