Before I really get started, let me just clear something up first: When we talk about verb “tenses,” we’re probably talking about a few related concepts: tense, mood, and aspect. In a nutshell, the most common distinctions are as follows:
Tense: The time something takes place, such as past vs. present (or really just non-past). For example, “walked” or “did walk” vs. the default “walk(s).”
Mood: The reality of the action or state, such actual vs. hypothetical, conditional, necessary, etc. For example, “walk(s)” vs. “would walk” or “must walk.” A generic “irreal” mood can also be used to talk about the future since nothing in the future has actually happened, so it’s all sort of imaginary.
Aspect: The duration of the action, such as punctual (a.k.a. “unitary”) vs. non-punctual (indicating continuous or repeated action). For example, “walk(s)” vs. “am/are/is walking.” Note that some verbs (known as “statives”) tend to lack this distinction and just take the default (punctual) form. For example, “know,” “want,” “own,” etc.
As for how such modifiers are applied to make new verb forms (commonly just called “tenses”), all sorts of variation is possible, but the most natural pattern seems to be to follow this order of tense, mood, aspect, and then finally the main verb. So in the creole language of Hawaii, you’d have something like “He been go stay walk,” to mean “He would have been walking.”
Now that all that groundwork’s out of the way, let me tell you about a pattern I’ve observed in various languages and how you can use it to build a consistent, flexible verb system from the ground up. Instead of using the word “will” to indicate a future action in English, it’s extremely common to use a form of “be going to” before the main verb. You can do something similar with most so-called “modal verbs” (“must” → “have to,” “can” → “be able to,” “should” → “be supposed to,” etc.).
What makes this pattern desirable when you’re designing a new constructed language is that it lets you get away with fewer “parts of speech,” meaning you can be more consistent with how words are allowed to combine. This makes the whole thing easier to learn with fewer arbitrary grammatical rules, which makes everybody happy.
Think of all the different shades of meaning you can get by using different vocabulary before the main verb. Instead of “I’m going to walk,” you could say “I’m about to walk.” Or you could add on extra descriptive detail, as in “I’m just barely able to walk.” Getting away from grammatical bits and bobs like function words, particles, and affixes means you can focus on building up your language’s vocabulary, which is honestly the only part that most people are going to care about anyway.
Following this approach, how can we communicate the basic tense/mood/aspect distinctions we covered earlier?
Tense: Following the pattern of “be going to” to indicate the future, how about “be coming from” to indicate the past? So “I will eat,” becomes “I’m going to eat,” and “I ate,” or “I did eat,” becomes “I’m coming from eating.” This seems like quite a mouthful just to indicate the past tense, but if you simplify things just a bit, these forms could be more like “I go eat,” and “I leave eat.” The vocabulary is up to you, so be sure and make the most common words really short.
Mood: Probably the most generic irreal modal in English is “would,” so let’s tackle how this might be expressed using verb stacking like we did for tense. “I would eat,” might be recast as something like “I’m imagined to eat.” So the actual verb we’d want to indicate an irreal mood would have an English translation of something like “be imagined to.” Simplifying this again using only English monosyllables, we’d have something like “I thought eat,” where “thought” would indicate a passive voice version of “think,” not actually past tense, which would instead be marked with “leave.”
Aspect: Following the example of Hawaiian Pidgin noted earlier, we could indicate a non-punctual aspect using the word “stay,” so “I’m eating,” would become “I stay eat.” Or we could use the word “keep,” as is often done in standard English already, but it might make more sense to save that for repeated actions.
Putting it all together, “I would have been eating,” becomes “I leave thought stay eat.” It’s kind of a mouthful, but so is the standard English equivalent, and stacking ordinary verbs to mark tense, mood, and aspect gives you way more expressive power in the long run. To further improve this system, let’s look back at the “be going to” construction we started out with.
Of course, “be going to” is commonly shortened down to “be gonna,” particularly in speech. Going even further with this, a phrase like “I am going to” can be contracted all the way down to “I’ma.” The thing to note here is that even if it looks like you’re stacking up a bunch of separate words and you’d like things to be more compact in your language, you can always fuse common sequences together like this to form new words. That’s an extremely common practice in natural languages, even if it isn’t always reflected in writing. In fact, this is a pretty likely explanation for where the different verb endings in languages like Latin came from in the first place.
So if you were to construct an English-based creole language as I’ve sketched out here, you might wind up with the following simplified auxiliary verbs (for a start):
go = future tense, “will,” “to be going to”
liv = past tense, “did,” “-ed,” “to have done”
tot = hypothetical/conditional mood, “would,” “to be imagined to”
nid = “must,” “to have to,” “to need to”
eb = “can,” “to be able to”
spos = “should,” “to be supposed to”
ste = continuous aspect, “-ing,” “to be doing”
kip = repeated aspect, “to keep doing”
And from these building blocks, you’re free to form all the contractions you want to make your language as naturalistic and irregular as you can stand. Go from “I would have been eating,” to “Mi liv tot ste it,” to “Mi li’to’se it,” all the way down to “Mi los it,” if you like. Have fun!