Os novos céticos transmitiram estes cinco modos da suspensão do juízo [epoche]. [...] O primeiro diz respeito ao desacordo: verificamos que, quanto a uma proposição que nos põem sob os olhos, há na vida e nos filósofos um desacordo que se não pode evitar; e, por consequência, não sendo possível preferir ou contestar, chegamos à suspensão do juízo. O segundo é a regressão até ao infinito: dizemos que a prova que aduzem para garantir uma proposição tem necessidade de uma outra prova, e esta de outra, até ao infinito; por isso, como não sabemos por onde principiar o raciocínio, a suspensão do juízo é a consequência natural. O terceiro é extraído da relatividade: o objeto aparece-nos tal ou tal consoante aquele, que julga e segundo as circunstâncias da observação, mas abstemo-nos de julgar aquilo que existe por natureza. O quarto modo é o do postulado ou da posição de base: contestados até ao infinito, os dogmáticos consideram um ponto de partida que não demonstram, mas o qual julgam digno de obter o seu assentimento de modo absoluto e sem demonstração. O quinto modo é o do círculo vicioso: aquilo que deve confirmar a coisa em questão precisa de ser demonstrado precisamente pela coisa em questão; por isso, não podendo tomar nem um nem outro para encontrar o outro, abstemo-nos de julgar acerca de um e do outro. É possível reduzir a estes modos tudo quanto está em questão.
 The more recent skeptics hand down five Modes of suspension of judgment, as follows: first, the one from dispute; second, the one that throws you into an infinite regress; third, the one from relativity; fourth, the hypothetical one; and fifth, the reciprocal one.  The one from dispute works by our finding that an unresolved standoff is in place on the matter under consideration, both in ordinary life and among philosophers; because of this, we can’t choose or reject anything, and so we end up at suspension of judgment.  The one from infinite regress is where we say that what’s brought forward as a guarantee on the matter under consideration is in need of a further guarantee, and that one is in need of another, and so on to infinity, so that we don’t have anywhere from which to begin building things up, and suspension of judgment follows.  The one from relativity … is where the actual object appears this way or that in relation to what’s doing the judging (and the things observed at the same time), but we suspend judgment on how it is in its nature.  The one from hypothesis is when, on being thrown into an infinite regress, the dogmatists start from something that they don’t establish, but decide it’s OK simply to assume as a matter of agreement, without demonstration.  And the reciprocal Mode arises when what is supposed to provide confirmation of the matter being investigated is in need of a guarantee from the thing under investigation; and so, being unable to assume either for the purpose of establishing the other, we suspend judgment about both.
That it’s possible to bring every object of investigation under these Modes, we will show briefly as follows.  The point under consideration is either something perceived or something thought, and whichever it is, there is dispute about it; for some people say that it’s only things perceived that are true, others that it’s only things thought, and others that it’s some things perceived and some things thought. So, are they going to say that the dispute is decidable or undecidable? If it’s undecidable, we have the result that we must suspend judgment; for about matters of unresolved dispute it’s not possible to take a position. But if it is decidable, we ask on what basis it is going to be decided.  For example, will the thing perceived (we’ll set up the argument with this one first) be decided by a thing perceived or by a thing thought? If it’s by a thing perceived, then since things perceived are what our investigation is about, that too will need something else as a guarantee. And if that too is a thing perceived, it will also need something else to provide a guarantee, and so on to infinity.  But if the thing perceived has to be decided on by a thing thought, then since things thought are also a matter of dispute, this too, being a thing thought, will need to be decided on and guaranteed. Where, then, will the guarantee come from? If it’s from a thing thought, there will be a similar infinite regress; but if it’s from a thing perceived, then since a thing thought was taken on as a guarantee for the thing perceived, and a thing perceived as a guarantee for the thing thought, the reciprocal Mode is introduced.  But if, to avoid this, the person talking to us decides it’s OK to assume something as a matter of agreement, without demonstration, for the purpose of demonstrating the things that come next, the hypothetical Mode will be introduced—and that’s a dead end. For if the person giving the hypothesis is trustworthy, we will be no less trustworthy whenever we give the opposite hypothesis. And if the hypothesis the person gives is something true, he makes it suspect by taking it as a hypothesis instead of establishing it—while if it’s false, the foundation of the things being “established” will be unsound.  Besides, if giving a hypothesis gets us somewhere in creating trust, let’s give as a hypothesis the very thing being investigated—not something else by means of which the matter we’re arguing about is going to be established. But if giving as a hypothesis the thing being investigated is absurd, it will also be absurd to give as a hypothesis something higher up.10  And that everything perceived is relative is clear: it’s relative to the perceivers.
It’s obvious, then, that whatever perceived object is proposed to us, it’s easy to bring this under the Five Modes. And we reason in the same way about what is thought. If it is said to be an object of unresolved dispute, we have to suspend judgment about it—that will have been conceded.  But if the dispute is going to be decided, then if it’s by means of a thing thought, we’ll throw them into an infinite regress—and if it’s by means of a thing perceived, into the reciprocal Mode; for since what is perceived is in turn a matter of dispute, and cannot be decided by means of itself because of the infinite regress, it will need what is thought, just as what is thought needs what is perceived.  For these reasons, assuming something by hypothesis will again be absurd. And things thought are also relative: it’s relative to the thinker that they get their name, and if they were by nature such as they are said to be, they would not have been an object of dispute. What is thought, then, has also been brought under the Five Modes, which means that it is necessary all around for us to suspend judgment about the matter under consideration.