Table of Contents

Isn't it weird that out of everything in this article, this section isn't actually a table? Go ahead, inspect element. It's just a list. Also please don't look too closely at my messy code, I'm shy.

• 1: Preamble - An Analysis of P2TT from the Ground Up
• 2: Introduction to Phonetics
• 3: The Data Zone
  • 3.1: Phonetic Identification
  • 3.2: Transcripts
    • 3.2.1: Petscop 11
    • 3.2.2: Petscop 15
    • 3.2.3: Petscop 23
    • 3.2.4: Petscop Soundtrack
• 4: Analysis Time
  • 4.1: P2TT Itself
  • 4.2: Interjection: Analytical Perspectives
  • 4.3: De-Interjection: The Purpose(s) of P2TT
  • 4.4: The Big Three
  • 4.5: Mistakes and Other Anomalies
    • 4.5.1: A Near-Open Front Rounded Vowel By Any Other Name
    • 4.5.2: Other Things I Noticed
  • 4.6: Bonus Charts and Stuff
• 5:

1: An Analysis of P2TT from the Ground Up

Player 2 To Talk, henceforth abbreviated "P2TT", and also referred to as "the gamepad language" in the Petscop Fandom's foundational text the PCPD, is a mechanic in Petscop (game) within Petscop (webseries). It appears in four episodes as an in-game chat function used by three(ish) characters. It is used to communicate a few very important things, and several incidental details. The purpose of this article is to provide a complete documentation and exhaustive analysis of P2TT; the PCPD contains some inaccuracies and gaps in information, and I am a huge nerd both able and willing to correct those inaccuracies. And because I could and would, I should, and in fact must.

Before we begin: this article is built around the assumption that the reader has the Catrinity font installed to properly display IPA symbols inline with regular text. If you don't have this font (ie, you are on a phone) the tables and such should still be readable, but they might look kind of ugly.

You will know Catrinity is installed correctly if these two faces match:

😧

This article is also written with the assumption that you have watched Petscop in full and have a basic understanding of its plot, characters, and themes. I'm not going to pause mid-paragraph to explain who or what I'm referring to. If you are, for some reason, only interested in the phonetic writing system part of this, I'm afraid you are going to have to wait for someone else to write the inverted version of this article, which has a crash course on the fundamentals of Petscop at the start followed by extremely in-depth commentary on phonology. And then, having reached the joining point of both phonetics and Petscop analysis, we will be able to syncretize the ultimate truth of both fields, and achieve world peace and stuff. So someone should probably get going on that. Not me though, I'm busy writing all this shit about analytical frameworks and making HTML tables and stuff.

2: Introduction to Phonetics

This section is a brief overview of phonetics as relevant to Petscop. (What a sentence that is.) If you're already familiar with the subject through exposure to or membership in the conglang community, langblr, etc., you can skip this. Otherwise:

Phones are units of language sounds. They are categorized by what parts of the mouth are involved in making them and how. This Tom Scott video is a good, quick explanation of how that system works, as well as an introduction to the International Phonetic Alphabet (IPA), the ubiquitous way to write phonetics. IPA symbols are going to show up throughout this article, so feel free to keep this Wikipedia page open in another tab for quick reference, though I will try my best to explain stuff here in the article. (Unless you have a really narrow screen resolution, you can probably see me starting to do that with a table a few paragraphs down!)

Because the pronunciation of individual sounds can vary a lot, phones are grouped into larger phonemes. A phoneme is a collection of all the sounds that speakers of a language or dialect perceive as "the same sound". Whether or not two sounds are "the same" depends on which language or dialect the speaker is most familiar with. For example, in my particular accent (Pacific Northwestern), the words "marry" and "merry" are homophones, but in Received Pronunciation (Southern England) those vowel sounds are heard as two different phonemes. It is very hard for me to hear the difference in pronunciation in a recording of someone with an RP accent saying these words. Keep this concept in mind when looking at the transcriptions of words in this article. There is no definitively correct way to pronounce any given word; the "right" way to pronounce something is however makes it understood by the person you're speaking to.

Back to the IPA: do you still have that Wikipedia page open? Don't worry about memorizing exactly what symbol corresponds to what sound. If you have an interest in phonetics or phonetics-adjacent fields, you will learn this through exposure or already have. If you don't care about it at all outside the context of Petscop, you will not need to. I'm going to break it down as far as you need to know, and you are free to forget about it immediately after.

Most of the consonant symbols are pretty intuitive: /m/ goes "mm", /k/ goes "kuh". The few that aren't as simple are generally for sounds that are written with digraphs, plus /ɹ/. Don't worry about /ɹ/, it's just a normal "arr" sound. It's upside down because there are a million different ways languages make rhotic (r-like) sounds and the right side up "r" is already used for a different one. Don't worry about the other sounds either, because I'm going to explain them right now.

/j/	makes a "yuh" sound, as in "hallelujah". when put between a consonant and a vowel, it means that the vowel starts with a "y" sound - the difference between "moo" /muː/ and "mew" /mjuː/.
/ʃ/	makes a "shh" sound on its own, or a "ch" sound as /tʃ/ - the difference between "shin" /ʃɪn/ and "chin" /tʃɪn/. notice that "ch" feels mostly the same as "shh" in your mouth, except that your tongue hits the ridge behind your teeth at the start: that's the /t/, the voiceless alveolar plosive, and that ridge is the alveolar ridge.
/ʒ/	hard to associate with any one letter. /ʒ/ is the voiced post-alveolar fricative, which means it is to /ʃ/ as /v/ is to /f/, or as /z/ is to /s/. /dʒ/ is a "soft g", as in "gentle", so /ʒ/ by itself is like a "soft g" without that little forward movement of the tongue at the start.
/θ/ /ð/	makes a "th" sound - but have you noticed something about the jargon words in italics in the cells above? voiced and voiceless describe whether or not you buzz your vocal chords when making a sound, and english has both a voiced "th" and a voiceless "th". this is hard to notice unless someone points it out, because there are very few word sets where the only differing phoneme is the voiced or voiceless "th" - try "ether" /iːθər/ and "either" /iːðər/. /θ/ is voiceless, /ð/ is voiced.
/ŋ/	makes a "ng" sound, as in "bring". fun fact: this symbol is called the engma.

Okay, now we have to get into the scary stuff: vowels. Vowels are a lot mushier than consonants, and a lot of the variation across accents and dialects happens in the vowels. You can't rely on your intuitive knowledge of English spelling to figure out what sounds IPA vowel symbols correspond to. Try to just throw all those preconceived notions out the window, or squash them down underneath a paving stone or something, and think of these IPA symbols as being their own totally unique things which just happen to sometimes also resemble English letters.

/ɑː/ /ɒ/ /ɔː/	in my accent, all three of these phones are merged, and they all make the same "aahh" sound. P2TT merges /ɑː/ and /ɒ/ but not /ɔː/. in that case, /ɑː/ and /ɒ/ are "aahh" and /ɔː/ is "aw" - the difference between "all" /ɑːl/ and "awl" /ɔːl/.
/æ/	makes a short "ah" sound. /ɑː/ has your lips relaxed, and /æ/ has them pulled back - the difference between "cal" /kæl/ and "call" /kɑːl/. the best way to get this one down is just to spend a couple seconds making "aahh, ah, aahh, ah" noises at your computer screen. go ahead, i'll wait.
/aɪ/	makes a "long i" / "eye" sound. do not get tripped up by english "-ain" words like "pain" and "grain" - those are actually /eɪ/.
/aʊ/	makes an "ow" sound - the difference between "call" /kɑːl/ and "cowl" /kaʊl/.
/ɛ/	makes a "short e" / "eh" sound.
/eɪ/	makes a "long a" / "ayy" sound - the difference between "pen" /pɛn/ and "pain" /peɪn/.
/ɪ/	makes a "short i" / "ih" sound.
/iː/	makes a "long e" / "ee" sound.
/oʊ/	makes an "oh" sound - the difference between "got" /gɒt/ and "goat" /goʊt/.
/ɔɪ/	makes an "oy" sound - the difference between "so" /soʊ/ and "soy" /sɔɪ/.
/ʊ/ /ʌ/	in my accent, these are merged and make the same "uh" sound. P2TT distinguishes between them, using /ʊ/ for "put" and "could" and /ʌ/ for all other "uh" sounds.
/uː/	makes an "ooh" sound - the difference between "sun" /sʌn/ and "soon" /suːn/.
/ə/	this symbol is called the schwa, and the sound it makes is hard to pin down. it's in the middle of all the other vowels (the mid central vowel.) in english, it's mostly like an unstressed "uh" /ʌ/, but basically it can be whatever your heart desires.
/ɚ/	ok, one last thing, and this is the scariest one. this is an r-colored schwa, which is an example of a rhotacized vowel (vowel that is also an r) / a vocalic r (r that is also a vowel.) when a word ends in "-er" and you pronounce it by just sort of lowering your voice instead of making the full /ɹ/ sound, that's /ɚ/. it is functionally the same as /əɹ/, but P2TT condenses it into 1 button press instead of 3. gotta save those frames, am i right.

Congratulations! You made it through the phonetics lesson. Now we can actually talk about Petscop. Remember Petscop? You started this article because you wanted to see me analyze some Petscop, and I tricked you into learning what a voiced post-alveolar fricative is. What the hell.

3: The Data Zone

3.1: Phonetic Identification

In P2TT, inputs of one or two buttons are interpreted as phonemes. Input 1 can be any of the trigger buttons, or empty, and input 2 is any of the directional / shape buttons or START. Presumably, the SELECT button is used to finalize a message, because it is the only button not used in the phonetic table. The following table shows which inputs are correlated to which phonemes.

					empty
	/ŋ/	/ʌ/	unused	/θ/	/ɛ/
	/dʒ/	/j/	/oʊ/	/f/	/aɪ/
	/g/	unused	/ʊ/	/ð/	/ɚ/
	/k/	/h/	unused	/v/	/æ/?
	/z/	/n/	unused	/p/	/æ/
	/ʃ/	/l/	/ɪ/	/b/	/ɔː/
	unused	/ɹ/	/iː/	/t/	unused
	/s/	/m/	/eɪ/	/uː/	/ɑː/
	/tʃ/	/w/	unused	/d/	/ə/

3.2: Transcripts

Below I have transcribed every instance of P2TT seen on-screen in the series. Some notes:

• The background color of each row indicates who said each line: green for Marvin, red for Paul, and purple for Belle/Tiara.
• Inputs marked with a question mark ? went out of frame and can't be conclusively identified. The presumed transcription is marked with (parentheses). If there is more than one likely answer, the possibilities are separated with question marks ?.
• Transcriptions written in italics and marked with an asterisk * are ones I think are particularly interesting and will discuss in more depth in the Analysis section.

Petscop 11

INPUTS	PRINT	TRANSCRIPTION
	Hello	hɛloʊ
	Pall	pɔːl
	Hell	hɛl
	Hello	hɛloʊ
	Funny	fʌniː
	Ha	hɑː
	Ha	hɑː
	Play	pleɪ
	Music	miːuːzɪk *
	For	fɔːɹ *
	Baby	beɪbiː
	She	ʃiː
	Will	wɪl
	Become	biːkəm
	Melody	mɛlədiː
	Thanks	θænks
	Lovely	ləvliː
	She	ʃiː
?	Tripped	tʃɹɪp(d?t?) *
?	And	eɪn(d) *
	Fell	fɛl
?	And	eɪn(d)
	Is	ɪs *
?	Lost	lɔːs(t)
	Stop	stɑːp
	Sorry	sɑːɹiː
	Tiara	tiːɑːɹə
	Plays	pleɪs
	Bad	bæd
	Music	miːuːzɪk
	Too	tuː
	Do	duː
	It	ɪt
	Right	ɹaɪt
	Next	nɛkst
	Time	taɪm
	Sad	sæd
	OK	oʊkeɪ
	Pall	pɔːl

Petscop 15

INPUTS	PRINT	TRANSCRIPTION
?	Not In Table	mɑːɹvɪ(n) *
	Sit	sɪt
	Here	hiːɹ
	For	fɔːɹ
	The	ðʌ
	Present	pɹɛzɛnt
	This	ðɪs
	Is	ɪs
	Bell	bɛl
	I	aɪ
	Am	æm
??	Tiara	tiːɑː(ɹə)
	Not	nɑːt
	Bell	bɛl
?	Press	pɹɛ(s)
???	Nifty	nɪf(tiː)
	What	wʌt
	Nifty	nɪftiː
	No	noʊ
?	Player	pleɪ(ɚ)
	One	wʌn
?	Press	pɹɛ(s)
???	Nifty	nɪf(tiː)

Petscop 23

INPUTS	PRINT	TRANSCRIPTION
	Pall	pɔːl
	What	wʌt
	Room	ɹuːm
	Are	ɑːɹ
	You	juː
	In	ɪn
	Thanks	θænks
	Here	hiːɹ
	I	aɪ
	Come	kəm
	What	wʌt
	No	noʊ
	No	noʊ
	Not In Table	dæ *
	I'm	aɪm
??	Sorry	sɑːɹ(iː)
	Help	hɛlp
	Put	ʊt *
	Baby	beɪbiː
	Time	taɪm
	Too	tuː
	Play	pleɪ
	Fuck	fʌk
	Stop	stɑːp
	Stop	stɑːp

Petscop Soundtrack

INPUTS	PRINT	TRANSCRIPTION
	There	ðæɹ *
	Is	ɪs
	Boss	liːnə *
	Waiting	weɪtɪŋ
	For	fɔːɹ
	Her	hɚ
	Son	sʌn
	Pall	pɔːl
	Do	duː
	You	juː
	Remember	ɹiːmɛmbɚ
	Being	biːɪŋ
	Born	boʊɹn
	Smuggled	smʌgld *
	Away	ʌweɪ
	Driving	dʒɹaɪvɪŋ *
	Too	tuː
	Your	joʊɹ
	New	nuː
	House	æuːs *
	Boss	liːnə
	In	ɪn
	Driver	dʒɹaɪvɚ
	Seat	siːt
	Me	miː
	In	ɪn
	Back	bæk
	I	aɪ
	Could	kʊd
	Not	nɑːt
	Wait	weɪt
	Too	tuː
	Be	biː
	Your	joʊɹ
	Friend	fɹɛnd
	Family	fæmɪliː
	We	wiː
	Can	kæn
????	Investigate	ɪnvɛstɪg(eɪt)
	This	ðɪs
	Together	tuːgɛθɚ

4: Analysis Time

4.1: P2TT Itself

The P2TT system has 45 possible inputs, of which 38 are used on-screen. There are apparently two inputs for /æ/ (the nuances of which will be discussed in a later section), making 37 phonemes in the system as witnessed. There are approximately 44 English phonemes. The major ones not seen in Petscop are:

/ʒ/	our friend the voiced post-alveolar fricative, used in words like "beige" /beɪʒ/ and "measure" /mɛʒəɹ/.
/aʊ/	"ow!"
/ɔɪ/	"oi!"
/ʔ/	this is the glottal stop, which i normally would not consider a "major" english phoneme if not for the presence of the word "uh-oh" /ʌʔoʊ/ in petscop.

So, if you wanted to make a fully compatible P2TT table for writing secret messages to your friends, my recommendation would be to fill it in like this:

					empty
	/ŋ/	/ʌ/	/ɔɪ/	/θ/	/ɛ/
	/dʒ/	/j/	/oʊ/	/f/	/aɪ/
	/g/	/ʒ/	/ʊ/	/ð/	/ɚ/
	/k/	/h/	/ʔ/	/v/	/æ/?
	/z/	/n/	free space	/p/	/æ/
	/ʃ/	/l/	/ɪ/	/b/	/ɔː/
	free space	/ɹ/	/iː/	/t/	/aʊ/
	/s/	/m/	/eɪ/	/uː/	/ɑː/
	/tʃ/	/w/	free space	/d/	/ə/

You can use the remaining three spaces for r-vowels of your choosing, marginal segments like /x/, blowing raspberries, the "I don't know" sound, other things that would start reddit fights about whether or not they're phonemic, clapping, the vine boom sound effect, whatever. Recall what I said about the only "correct" way to communicate being whatever is understood by your audience.

4.2: Interjection: Analytical Perspectives

There are some concepts I want to establish in your mind before we get mired in the theorycrafting. You might have heard of the "Watsonian-Doylist" dichotomy before. A Watsonian analysis looks at a story from an interior perspective, and explains things based on the world as it exists to the characters. A Doylist analysis takes an exterior perspective, and explains things based on the "real" world, or the world as it exists to the author. Specifically in the context of errors or discrepancies, the W/D framework determines whether something was a mistake on the character's behalf or the author's -- if a fictional character says the boiling point of water is 200°F, did they fuck up, or did the author?

The W/D framework is named after Sherlock Holmes' sidekick and creator respectively. 130 years later, stories have gotten somewhat more complicated, and characters and authors often interact with or are each other. In the case of Petscop specifically, there are layers of authorship and authored entities that are very confusingly all called Petscop. Here's a diagram I made.

presented in cheerful seussian easter egg colors, for theme

If there is something in Petscop that looks like a mistake, we as analysts must decide where in the chain the mistake happened. Let's presume that mistakes are visible to everyone up-chain in a linear way: Rainer's mistakes are visible to Rainer, Paul, The Family, and Us in the "real world"; Paul's mistakes are visible to Paul, The Family, and Us; The Family's mistakes are visible to The Family and Us; and Tony Domenico's mistakes are visible only to Us. If Tony spells something wrong in a caption -- assuming captions are non-diegetic accessibility tools -- Paul doesn't know or care about that, because he is a fictional character who cannot watch youtube videos.

For the subject of P2TT, because it is an in-game system, the only real options for authorial culpability are Rainer or Tony. Paul and The Family don't have control over how P2TT operates, as far as we know. Some quirks of P2TT seem to be cut and dry Domenican mistakes, and some could be explained either Domenicanly or Rainerianly. I will present both options when possible.

That's as well as I can prepare you for the rest of the article. From here on out, I am going to assume you know the meanings of all the IPA symbols, as well as of my funny made-up words highlighted in green and yellow.

4.3: De-Interjection: The Purpose(s) of P2TT

From a Rainerian perspective: P2TT is a way for players to talk in real(?)-time. The multiplayer functions of Petscop are obtuse, and player interaction might be based on demo replays or other Weird Petscop Shit (WPS). As is typical of WPS, P2TT is somewhat hostile to the player. It is slow to use, unoptimized (some common phonemes take multiple buttons while rarer ones hog the single-input slots), has a limited dictionary of words, doesn't understand context-sensitive homophones, and hogs the second controller slot that ostensibly could've been used for an actual Player Two, eliminating the need for this system in the first place, because you could just talk to your coplayer with your mouth. Children often spell words as they sound, but any advantage this could harbor for players is undone by having to learn formal phonetics and an entirely new way to write them. There's no T9-esque system to help with long words, so players have to make every single input, which is usually longer than the word as written in English. "Investigate", an 11-letter word, takes 19 inputs in P2TT. The only thing P2TT has going for it is that it doesn't block the screen like the keyboard prompt for asking questions of the TOOL.

From a Domenican perspective, P2TT makes a lot more sense. It's a way for Paul to talk directly to other characters in a way visible to the audience. Excepting the one part where we can kind of barely hear Belle on the other end of the phone, Paul is the only speaking character in Petscop. P2TT and Draw Mode are the only forms of communication besides the one-way dialogue boxes.

P2TT also evokes the forms of "easy chat" or "safe chat" systems in kids' virtual worlds, where messages were composed by picking from premade words or phrases in order to stop kids from ERPing or posting their social security numbers. This is probably a nostalgic memory for most of the audience, and Petscop uses that to reinforce the tone of the series by allowing P2TT to render "hell" and "fuck" -- it's more subtle, because Paul doesn't say anything about it when it happens, but it serves the same function as the iconic "that's a dead kid" and "I'll shoot her in the head" moments. Petscop manages a very, very controlled dosing regimen of creepypasta scares between weird-but-not-frightening gameplay. As described on my Fun Links page, it really is the distilled essence of every haunted video game from the golden age of creepypasta. ...Not that swear words are particularly scary; just that they remind us that this game, in its current form, is not actually for children, and things that look like they might be for children but aren't are creepy.

I also think P2TT resembles some of the other janky text-input systems from contemporary video games, like Dōbutsu no Mori for the N64.

source: N64 Archive

I know this kind of system is somewhat common in Japanese games and programs, but I played an English patch of this and it was nigh unusable. I mentioned T9 earlier, and I was in probably the last cohort of people who actually learned how to use it -- my first phone was a double-hand-me-down Super Slice in 2010 -- though I never used it in a video game (the only one I had was Ms. Pac-Man and I never got a high score.) For you youngsters out there, T9 is the system whereby you press the numbers on a touch-tone dial pad a bunch of times in a row to select letters and it uses lookup tables to guess what word you're typing. Here's someone else's article with some more examples of other gamepad input systems of past and (as of 2017) present. As one more tangentially related personal anecdote: I used to browse cheezburger.com on the Wii with the keyboard from my parents' Bondi Blue iMac G3 plugged into the USB slot. Now that's a fucking sentence.

What I'm saying is that P2TT is not that far removed from the other weird inputs people were dealing with in the 90s/00s, and it's fun to imagine a world where it caught on and kids got really good at button mashing phonetic transcriptions of words.

P2TT is also a way to hide secrets, which is fun for the author and the audience. There are three P2TT instances that contain meaningful secrets, which will be discussed below. I'm warning you now that everything after that is completely dorky phonetics-themed cornplating. If you choose to read it anyway, you will be rewarded a/o punished with more charts and tables.

4.4: The Big Three

The three P2TT anomalies that are unambiguously intentional storytelling devices are:

	INPUTS	PRINT	TRANSCRIPTION
1.	?	Not In Table	mɑːɹvɪ(n)
2.		Not In Table	dæ
3.		Boss	liːnə

1. In Petscop 15, Paul tries to say Marvin's name, but P2TT doesn't parse it as a word. P2TT apparently only has one person's name in its dictionary, which is special case number 3. Other names are interpreted as homophonous common words, "Pall" and "Bell", or are already common words in the case of "Tiara". It might not be a coincidence that "Care" and "Carrie" ("carry") would also pass through P2TT. Why the system didn't include special features for the names of the kids is unknown. As a multiplayer feature, it would have been pretty useful to be able to say your coplayers' names. Some of the other Petscop Kids' names are common words or homophones, but the rest of them would be out of luck.

Because I am addicted to making tables, here are the names of all the Petscop Kids and whether or not they would be allowed by P2TT. Names in green are already words, names in yellow have obvious homophones, names in orange are edge cases based on P2TT's messy phonemes, and names with no color would probably not pass the check.

Names
Adam	Amber	April	Belle
Ben	Care	Charlie	Daisy
David	Doug	Elly	Emily
Fatima	Henry	Holland	Jack
James	Jessica	Joel	Kate
Kyle	Larry	Laura	Lucas
Mackenzie	Meghan	Melinda	Mike
Nathan	Nick	Paul	Peter
Phil	Ramona	Sally	Sarah
Sean	Shelby	Tiara	Theo
Tony	Trevor	Welles	Will

("Ben" depends on if P2TT parses /bɛn/ as "been".)

"?" is the first Not In Table input, and it tells the audience that P2TT will not print words it doesn't recognize, something that might have been inferred by how Paul's name is rendered as "Pall" but is now made explicit. This is setup for the second Not In Table input, which is much shorter but more important.

2. In Petscop 23, Paul starts to say something, but apparently doesn't finish typing. The input hangs for 100 seconds, during which there is a black cut scene possibly indicating that more time has passed. After Marvin leaves the room, either the input times out or Paul finally sends the message without finishing the word.

I think we all know what he was trying to say, and it has huge implications for the entire series.

Ok sorry I'll stop fucking around. The pretty obvious reading is that Paul is about to say "dad". As a reminder, the surrounding lines are:

INPUTS	PRINT	TRANSCRIPTION
	Here	hiːɹ
	I	aɪ
	Come	kəm
	What	wʌt
	No	noʊ
	No	noʊ
	Not In Table	dæ
	I'm	aɪm
??	Sorry	sɑːɹ(iː)
	Help	hɛlp

This scene is one of the most important in Petscop. The 100 seconds between when Paul stops typing and when he comes back are so incredibly tense. In this moment, the events of Petscop The Series are unequivocally extending beyond what we can see in Petscop The Game. This confirms that the rooms are real places that Paul is really in, and that Marvin really has access to, and if this is real, then it gives validity to other parts of the game that we might not have been sure of. Things like the Child Library are still pretty clearly just game abstractions (nobody has tried, but it is probably impossible to build an infinite tower of bedrooms controlled by a magic easel), but because the testing rooms are real places, we become more willing to believe that the school, the house, the roadway, are also real. And, most importantly, that the potential violence enacted in these places is also real.

Petscop leaves a lot to the imagination. This is an extremely intentional choice, as evidenced by the censor bars imposed in-universe by The Family; information is plainly obviously redacted, so that the lack of information itself becomes information. Sometimes the shape of that information can be sleuthed out from tiny parts that are left uncovered -- remember, we're looking from a Domenican perspective right now, everything The Family misses is constructed on purpose by the author -- and sometimes all we have are broader themes and context clues. This scene in Petscop 23 is about as broad as it gets. Literally nothing is happening on-screen. It's up to the audience to discern what that nothingness means. A named absence becomes a presence; whatever you decide happens in those 100 seconds is part of the story.

I try not to be overly squeamish in my media analysis, but, like. this is an article about the secret phonetic code that shows up in four episodes, you know? My opinions on P2TT are serious, but no level of seriousness I can apply to the discussion of long versus short vowels will let that topic stand on the same level as familial violence and abuse. It is, as they say, inacessible from below. So I'm not going to talk about what I think happened in Petscop 23, not right here and now. Absolutely not where you can maybe still see the damn daniel gif. The broader analysis of what Paul calling Marvin "dad" means is also going to have to wait for a hypothetical full-series writeup. This is the P2TT article.

The utility of P2TT in this scene is twofold. 1: it provides suspense by having the unfinished message hang. Recall how P2TT's sole advantage over the keyboard menu is that it doesn't cover up the screen. Although hiding the whole screen behind a menu would also be suspenseful, I think showing that nothing is happening in the game is more effective. It makes it unambiguous that Paul and Marvin's interaction is happening in the real world. 2: it obscures information, which is a critical part of how Petscop interacts with its audience. Understanding the story requires the viewer to put in some effort looking for secrets. P2TT is a flag that says "there's something here, but you're going to have to work for it." Petscop is not, technically, an ARG, as the audience has no input over what happens in the videos themselves -- they're all prerecorded media -- but because so much of the story relies on interpretation, every person's experience of it is unique. The value of secrets is that of ergodic literature:

The ergodic work of art is one that in a material sense includes the rules for its own use, a work that has certain requirements built in that automatically distinguishes between successful and unsuccessful users.

I don't know if it's really possible to fail a piece of media, but I think you can definitely fail to engage with it as intended. In some cases breaking the rules actually makes a work way better -- but usually it doesn't. As, ostensibly, a puzzle game, it's not surprising that Petscop's rules include solving all the little secrets it so obviously presents us.

3. The final episode of the series, Petscop Soundtrack, is itself a trivial puzzle, because it's not labeled as a normal episode, and the story content happens undeclared at the very end of a 44 minute video that otherwise contains very little video content. Petscop's finale is, essentially, a monologue from Belle, during which one extremely important detail is masked in P2TT. When Belle says "Boss", her inputs actually say "Lina".

INPUTS	PRINT	TRANSCRIPTION
	Boss	liːnə

This is the one instance where the P2TT system makes special allowance for a person's name, and it does it by turning it into a different word. The reasons behind this are unknowable and all theories are complete speculation.

• Rainer (or whoever programmed the P2TT system) wanted to be able to refer to Lina without saying her name directly; there is/was some superstition about saying or writing Lina's name at the time the P2TT system was programmed, which Rainer later broke when he turned the game into his manifesto.
• Some form of emergent consciousness inside the game prevents Lina's name from being said by manipulating the P2TT tables. (But if that's true, why not just let it fall into Not In Table?)
• Boss is just a nickname Lina used and preferred over her given name.
• Rainer really, really doesn't know how to pronounce the word "boss"?

If you weren't paying attention to the details of P2TT before, it would be easy to assume that these inputs literally say "boss". You already saw that P2TT doesn't print proper nouns back in Petscop 15, and there's no precedent for it flat-out replacing one word with another. The fact that Belle is actually saying "Lina" is, if you'll allow me to be a little corny, a reward for following the rules: paying attention and solving puzzles. In this way P2TT acts as a sort of litmus test for people who are watching Petscop "successfully". If you can prove that you're really in the top percentage of Petscop understanders, you can be trusted with the information that Lina is the one waiting for her son Paul. What this means in the wider scope of things is up to you. P2TT is just the gate you have to pass through to get it.

4.5: Mistakes and Other Anomalies

That's the end of serious analysis time. Coming up now: these are the P2TT discrepancies that I think are pretty clearly just mistakes.

INPUTS	PRINT	TRANSCRIPTION	NOTE
	Music	miːuːzɪk	The /ju/ sound is being approximated with /iːu/ ("ee-ooh").
	For	fɔːɹ	"For" can be pronounced a lot of different ways, but I would not expect "fawr".
?	And	eɪn(d)	I would also not expect "eynd".
	Is	ɪs	"Is" is pronounced with a voiced alveolar fricative, /ɪz/.
	Put	ʊt	This input is missing the /p/ sound at the start ().
	There	ðæɹ	Unless you are pretending to be a pirate, "there" is pronounced /ðɛɹ/ or /ðɛɚ/.
	Smuggled	smʌgld	There should probably be at least one /ə/ in between those consonants at the end. Maybe P2TT is just really flexible about that kind of stuff? But probably not.
	House	æuːs	This is the most mangled pronunciation in the system -- trying to say this as written gives you something like "ah-oose". It seems like this is an attempt to approximate /aʊ/ with other vowel sounds. I don't know why this would be necessary, because at this point there are still eight unused button combinations that could be allocated to /aʊ/. I also don't know why there wouldn't be a /h/ at the beginning of this word, regardless of what vowels you wrote it with.

Mistakes are fine. These aren't load-bearing elements of the story, and I'm only pointing them out because I have an interest in phonetics. I don't expect someone who is already very busy making a series as involved as Petscop to be triple-checking their phonology. As it is, these are just incidental details for highly invested fans who have run out of other content to think about.

4.5.1: A Near-Open Front Rounded Vowel by Any Other Name

It is finally time talk about /æ/. Here's a quick table of places where it's used:

INPUTS	PRINT	TRANSCRIPTION
	Thanks	θænks
	Bad	bæd
	Sad	sæd
	Am	æm
	Thanks	θænks
	Not In Table	dæ
	There	ðæɹ
	House	æuːs
	Back	bæk
	Family	fæmɪliː
	Can	kæn

This seems to indicate that, according to P2TT, the vowels in "thanks", "bad", "sad", "back", "dad", and, confoundingly, "house", are all the same phoneme covered by the input. The vowels in "am", "there", "can", and the first vowel of "family" are likewise grouped under the input.

As stated above, "house" and "there" are probably just errors. "Thanks" is also a little strange; in my accent it's pronounced /θeɪnks/ (or /θeɪŋks/), and from what we can hear, Paul pronounces it the same way. At the very end of Petscop 8 he starts to say "thanks" and is cut off, but it definitely sounds like a long vowel, and in Petscop 14 (1:58) he says "thankfully" the same way. (A full investigation of Paul's speech patterns is beyond the scope of this article. Sorry if you were hoping for that.) /θænks/ isn't impossible or unheard of, so let's chalk it up to /æ/ () being faster to type than /eɪ/ () / a non-expert in phonetics misinterpreting that phoneme slightly. Not a big deal. That leaves us with the group of "bad", "sad", "dad", and "back", and the group of "am", "fam(ily)", and "can". This shows that whatever accent P2TT is based around (presumably WNE English, based on the (203) area code in Petscop 16) makes a phonemic distinction between /æ/ before nasal consonants and /æ/ before other consonants. I'm kind of working in the dark here, because these sound identical to me, and it looks like even the big shot phonologists aren't sure what exactly is going on here, but okay, whatever. Wikipedia tells me that the "tensed" version of /æ/ which occurs before nasals is usually transcribed as /ɛə/, so let's just go with that. is /ɛə/, is /æ/. It's also entirely possible that P2TT recognizes multiple pronunciations of the same word to account for differences in accents -- we don't see any examples of this on-screen, though, which implies either that it doesn't actually do that or that all speaking characters have the same accent. If it does group similar-enough pronunciations into the same word, then it doesn't really make sense to have a separate input just for these two extremely similar phonemes, but then again /ɚ/ also gets one all to itself, so who knows what they were thinking. I sure don't.

4.5.2: Other Things I Noticed

INPUTS	PRINT	TRANSCRIPTION
?	Tripped	tʃɹɪp(d?t?)
	Driving	dʒɹaɪvɪŋ
	Driver	dʒɹaɪvɚ

The only use of /tʃ/ in P2TT is in "tripped", and the only use of /dʒ/ is in "driving" and "driver". This is true to how these words are usually pronounced in American English: "chripped" and "jriver". I think it's a nice detail to have included this instead of just using /tɹ/ and /dɹ/.

INPUTS	PRINT	TRANSCRIPTION
	Together	tuːgɛθɚ
	Remember	ɹiːmɛmbɚ
	Family	fæmɪliː

When Belle says "together", she pronounces it as "too-gether", but when Paul says this same line all the way back in Petscop 2 (9:11) he pronounces it /təgɛθɚ/, "tugether". Belle also uses a long vowel in "remember", but Paul says that with a schwa too, as /ɹəmɛmbɚ/ (Petscop 2, 10:50). Evidently either Belle or the P2TT system itself is more careful about enunciation than Paul is. Paul does write "family" as a three-syllable word, even though he pronounces as two, /fæmliː/ (Petscop 14, 3:12).

4.6: Bonus Charts and Stuff

Pie Chart of Verbosity

Marvin and Belle use P2TT about the same amount, while Paul uses it vastly less than either of them. Paul is the only character to use P2TT in all four episodes it appears in, though.

Controller Heatmap

50	125	72	108	33	35	13	30	52	71	77	77	26

The right side of the controller is used a lot more than the left side. This distribution, if intentional, might mean that the player is meant to hold the P2 controller in their right hand and make most inputs with their right thumb, while holding the P1 controller in their left hand and using their left thumb to move their character with the D-pad. Or maybe you're supposed to use both hands for the P2 and move your character with your feet on the P1. As far as I'm aware, nobody has actually tried making their own canon-compliant P2TT input system with one of those USB controllers, so the most ergonomic way to use them is unknown.

If I knew like 12% more about programming, I would at least make a keyboard version. The most efficient schema I can imagine is WASD for the D-pad, OKL; for the shape buttons, L-shift/L-alt and R-shift/R-alt for the triggers, and space for start. Maybe double tap space to send? Kind of running out of easily accessible keys.

5: (Conclusion)

Have you ever felt a sense that either you needed to pierce into the core of something and comprehend every little bit of it, or else die?
(The Northern Caves, chapter 15.)

And that's everything I have to say about P2TT. Maybe this article will be useful to other Petscop fans, but even if it's completely unnecessary, I really enjoyed writing it. I will probably do something else in this vein eventually -- maybe a full-series look at Petscop, or another extremely granular analysis of passing details in another story, if there's one that catches my attention. Nothing is ever guaranteed, but I think it would be fun.

I'm not very concerned with reaching a large audience, but if you know someone who might be interested in this extremely niche topic, there is no reason not to share this post around. You can also message me on tumblr (or email me -- check the About page) with comments, corrections, suggestions, etc.

Beyond that: writing conclusions sucks and is mostly redundant anyway. Thanks for reading.