An Interview with Li Yi-Fan「小世界」訪談集 — 李亦凡

The below interview was published at Small World Journal, on the occasion of Taipei Biennial 2023. (scroll down for Chinese)

In the Taipei Biennial 2023, Li Yi-Fan’s work What Is Your Favorite Primitive employs a semi-autobiographical narrative method to explore the relationship between video production tools and people. How do the complex social and ethical issues arising from the interplay between users and makers affect the way we communicate? And how can software tools convey emotions that transcend individual feelings and perceptions? This work’s intimately detailed plot depicts human life and death, constructing a new politics of life with a unique aesthetic vocabulary. The artist spoke with curator Freya Chou about the influence of his childhood experiences on how he makes images and his subsequent creative path.

FC: The theme of this Biennial, “Small World,” is based on the concept of scale. Your work employs quite an extreme form of compression: you chose a production model very different from the large industrial scales required for films or video games. In your production process, you recreated all your digital moving images in your own studio. That seems incredible! How did you transform a collective, decentralized image production process into an individual DIY artwork?

LYF: When you first told me the exhibition would be about the concept of scale, that happened to be a subject I had been thinking about for quite a while: how I as an independent individual could fight the film industry. Actually, in the beginning I wanted to make movies. But I don’t like to bother other people. I’m very reserved. Even now, I’m a little embarrassed to ask my assistant to lend me a hand. So, my current situation actually comes from my background and the way I was raised.. And also, I’m a control freak. I like to handle things on my own. So when I pondered the possibility of a one-person film, I decided to harness the power of technology and use technological mass production to accomplish things that usually require a lot of manpower.

This thought process over the past two to three years led me to a change in mindset. At first, when I was producing a work, I found that I was endowed with enormous power—as if I was omnipotent. All I needed was to be able to write a software program, and I could do anything. But creating my work for the Taipei Biennial made me realize that actually I can’t escape large-scale manufacturing. In the past I used guerrilla tactics to hop from one industrially produced software program to another. So in What Is Your Favorite Primitive, it becomes more and more obvious what my relationship with these software programs or industries is—I seem to have all-powerful abilities, but in the end I still must comply with the rules of their production conditions. This made me think more deeply. I seem to have circumvented the multileveled hierarchy of filmmaking, but I was still stuck inside a set of specifications.

FC: That is to say, you attempted to alter the structure of production by resisting it in terms of form, but to a certain degree, you discovered that you ultimately couldn’t free yourself from big industry. It still surrounds you layer by layer.

LYF: My works always consider the struggle between cyborgs and naturalism. I could choose to reject all software and return to the state of primitive humans, to start from scratch with basic-level coding. But then I would also lose all the advantages of today’s technology.

FC: On first viewing I felt like the work might belong to a tradition of the grotesque or body horror. But then again, to an extent, the distortions of the body are presented with such calmness and clarity that one can feel visceral horror or even revulsion or disgust. How did you address the presentation of the human form? 

LYF: My starting point was very simple. I wanted to create motion within a digital environment. In the past, animation used key frames. You would keyframe every action one by one. But I don’t have the patience for that. I wondered if it was possible to create a performance using the more high-tech method of motion capture. But given my financial situation, I couldn’t afford to do a sophisticated performance. Only an animation studio with the entire gamut of technology can do that. So I used a very rough-hewn approach, with just a few basic pieces of equipment. The figures you see in my work weren’t really acted out by me; they were puppets I manipulated. And the weird bodily contortions actually came about because I was unable to manipulate the puppets with precision. Most puppeteers stand behind their puppets, but I stood in front of mine. This experience came from when I used to play with toys as a kid, and I would play with them face-to-face, inventing dialogues and performances by myself. That’s why it evolved into the deformed contortions you see today. But this deformity also accidentally resulted in a half-readymade, imperfect performative motion.

FC: So your current aesthetic expression arises from inferior technology. But paradoxically, at the outset you hoped to resist big industry by relying on your own powers, yet you discovered you were very limited and accidentally created a half-readymade aesthetic language. I think your entire creative process aligns well with the Taiwanese spirit. You started out feeling bashful, and due to a lack of material resources, you eventually created a “makeshift aesthetic.”

LYF: You could also call it “work around.” This is a specialized term in software programming. When you can’t solve something, you circumnavigate or piece parts together. This phrase is derogatory—resorting to the rough-hewn approach I talked about earlier, when searching for a solution.

FC: When I watched your video, I felt a rather jarring contrast between what feels like a very intimate, personal narrative and a visual language that I associate with the impersonal space of video game design or commercial animation. In what ways is this a personal, expressive work？Do we actually gain insight into you as a person?

LYF: My artworks have always started with myself. This may have something to do with how I grew up. I didn’t have any siblings, so I always played by myself as a child. Playing with toys as a kid is a form of narrative development, a conversation with yourself. So my art is basically an extension of my childhood experience of playing with Legos. But this personal experience is not important to viewers at all. I’m making use of technology to invite them into my world. It’s like I’m giving them a gift, and I’m enticing them to open it using beautiful packaging. The packaging is technology, and I’m wrapping my personal story in it. But for the viewer, the enjoyment doesn’t come from what’s wrapped up inside, but from the unpacking process.

FC: Sometimes digital art that includes human figures falls into the famous “uncanny valley.” Is that a territory that you aimed to explore with this work? 

LYF: The characters I create are me. It’s hard to separate the two. But lately I’ve often been asked if I use my art to deal with psychological issues or use it as an emotional outlet. At the moment of creation, I’m not thinking about it all that much. But after the fact, I think about it in more detail, and it seems this possibility does exist. I have a strong desire to express myself. And I make use of surrogates to help me process this desire, so I can achieve it in a relaxed way.

FC: I also want to raise a point that might seem counter-intuitive. This work feels strongly humanist to me. I say that because digital artists at this moment are often wrapped up in the financial world (crypto) or locked in some kind of struggle with AI (their work threatened with being made obsolete by this technology). Yet AI could never in a million years develop anything as authentically weird as what you’ve produced. And I would characterize the difference as “humanity.” It’s something that might explain why this work has connected so strongly with many viewers. If this makes sense to you, I wonder if you’ve thought about the difference between what’s “weird” about your work and the kind of “weird” that’s possible with AI? 

Li: That’s a huge question. For this Taipei Biennial, my work has already begun to delve at least partly into the relationship between AIartificial intelligence and image production tools, and my next work will explore it in even greater depth. I think this question comes back to the differences between humans and AI and their definitions. Lately, there’s been a lot of discussion about whether AI really understands language, since it possesses a language model and it can chat fluidly. This is a very major point of debate right now. But we could also ask, do we humans really understand language? Or are we just engaging in logical deduction like AI? At this moment in time, the difference between AI and me is a sense of humor. AI can’t perform logical reasoning about jokes yet, and that’s currently the biggest difference between humans and machines. So I really care about using humor to soften up some tough questions. That’s my own strategy.

FC: If AI could write jokes one day, I think that would be terrifying.

Li: Or if you saw a dog laugh at you with a sense of humor, if they learned a certain trait that is distinctly human, that would be really scary too.

FC: Can you reveal how the development of AI will figure in your next work?

LYF: I’ve always been curious about the tension between cyborgs and naturalism, which we mentioned earlier. I think AI is an unavoidable problem. In my work for this Taipei Biennial, I addressed this a little bit. When a tool contains AI, it includes complex specifications that constrain how you use that AI tool. This is a new concept for me: when a tool incorporates regulations for its own use. It’s like buying a hammer but no one tells you not to hit someone with it; instead, there is a “law” that keeps you from hitting people with the hammer. And yet, right now the trend in AI is for it to contain its own restrictions. My next work will be about the problem of regulating artificial intelligence, because it’s similar to the restrictions I’ve received about putting lurid content in my art. It’s as if my creative freedom is being regulated. But now the tools themselves engage in self-censorship. They directly prohibit you from producing lurid content. So my next work will focus on the relationship between AI and people. AI offers a lot of possibilities as a new creative tool, but many forms of surveillance and control have emerged within it that are invisible. I intend to start out from the perspective of the history of photography or the history of painting and explore the panics or prohibitions people faced back when those media were new.

FC: Did any other works in the Biennial resonate powerfully with you?

LYFi: I really liked Spencer Yeh’s 123 Trailers for Spectacle. The inspiration for my films comes mostly from B movies, so his work really struck a chord with me. And it stirred a new passion for cinema in me. I had gradually come to lose interest in movies, and it had been a long time since I had watched movies intensively. But his work reignited a desire to hunt down every one of those films and watch them. To a certain extent, we both had very similar starting points. B movies are images with a very strong sense of medium, because they don’t let you grasp the storyline right away—you have to keep waiting until the end. So they’re very good at manifesting the special characteristics of film as a medium. What makes 123 Trailers for Spectacle so interesting is that it uses other people’s materials, but it’s very punctilious about the pacing of the images and the musical accompaniment. It more or less reassembles other people’s works. It’s also a quite rough-hewn conceptual approach.

Translated by Brent Heinrich

李亦凡於2023台北雙年展展出的作品《難忘的形狀》，以半自傳的敘事手法探問影像生成工具與人之間的關係：在使用者與生產方之間所牽扯出複雜的社會倫理問題，是如何影響我們的溝通方式？而軟體工具又是如何能傳遞出超越個人情感認知的情緒？此作以私密的情節描述人類的生死殞落，透過獨特的美學語彙，構築出一套新的生命政治學。藝術家與策展人周安曼暢談了自小的生活經驗是如何影響他對影像的建構及往後的創作軌跡。

周安曼（以下稱周）：這次雙年展的主題「小世界」從尺寸規模的概念出發，就你的作品而言，具有一個相當極端的壓縮：你選擇了一個不同於電影或電玩產業需要大型工業規模下進行的生產模式，在你的數位流動影像製作過程裡，全部都是在自己的工作室中重新創建而成。這聽起來很不可思議！你是如何將集體、分散的圖像製作過程轉變為個人 DIY 的作品？

李亦凡（以下稱李）：當初你們跟我提及展覽是關於「尺度」的概念時，那剛好是我一直在思考的問題：如何以自身個體獨立對抗大型的電影工業。其實我一開始是想拍電影的，但我是一個不喜歡麻煩他人的人，我臉皮很薄，就算是現在有請助手幫忙，我還是會覺得不好意思。所以……現在的狀態其實是來自我的成長背景及訓練，就是所有事物都必須自己來。另一方面也是因為我是名控制狂，我喜歡自己有所掌握。所以在思考一人電影的可能性時，主要會是透過科技的力量，本來需要很多人力的事情，我藉由科技量化來處理。

過去兩、三年的思考過程帶給我一些心境上的轉變，從一開始製作作品時發現自己被賦予很強大的力量，好像無所不能，只要寫程式的能力夠，就什麼都可以做得出來。但直到北雙這件作品才讓我理解到，我其實無法擺脫大型的產業。我之前一直是以打游擊的方式遊走在產業出產的軟體之間，所以從《難忘的形狀》裡會越來越明顯看出我跟這些軟體或產業間的關係是什麼——好像我看似擁有無所不能的能力，但最終仍需符合他們生產條件下的規則。這也讓我進一步思考，我看似省去了很多層級的電影制度，但另一方面還是深陷在一個規範裡。

周：也就是說，你試圖透過形式上的對抗來改變生產結構，但某種程度上發現自己終究擺脫不了大型產業，還是被層層包圍住。

李：對我而言，我的作品一直在思考賽伯格對抗自然主義，我可以選擇拒絕所有軟體，回到原始人的狀態，很底層的codeing，從頭來過，但我也可能失去當今科技的優勢。

 周：一開始觀看這部作品時，會覺得你可能在討論怪誕或肉體恐怖的傳統。但某種程度上身體的扭曲卻又呈現得如此平靜和清晰，以至於會感覺到發自內心的恐懼，甚至是反感或厭惡。你是如何看待人體形態的呈現？

李：出發點其實很單純，我想在數位環境下創造動作。以前動畫使用關鍵影格（key frame），一格格key出每一個動作，但我是一個沒有耐心的人，我就想是否可以用動態捕捉這種比較科技的方式來表演。但以我的經濟能力是無法做出精緻的表演，那需要有整套完整技術的動畫工作室才能做到。所以我是用很土炮的方式，一些基本的設備來創作。作品中你會看到不是我在演那個偶，而是我去操作，而人體的奇怪扭曲其實來自我無法精準的操作人偶。一般的操偶師是站在偶的後面，但我是站在偶的前面。這經驗來自小時候玩玩具時，是我跟玩具面對面在玩，一種自我對話跟表演。也因此演變成現在看到的畸形扭曲，但這畸形卻又意外帶出一種半成品、不完美的表演動作。

周：所以現在的美學呈現是因為技術的不良，但弔詭的是，你一開始是希望靠一己之力來對抗大型工業，卻發現有很多侷限，因而意外創造出一種新的半成品美學語言。我覺得你整個創作都很符合台灣精神，從一開始因為「不好意思」，到後來因為物質匱乏而創造出「將就美學」。

李：也可以說是work around，這是寫程式的一個專有名詞，當你無法解決時就用繞道或拼湊的方式。這個詞帶有貶義，就是找方法解決，還是回到我之前說的土炮方法。

周：這次的作品讓我感到一種非常親密的個人敘事和一種視覺語言之間的鮮明對比，這種視覺語言讓我將與電玩設計或商業動畫的非個人空間聯繫起來。從哪些方面來說，這是一件個人的、富有表現力的作品？透過作品我們真的能深入了解你這個人嗎？

李：我的作品一直都是從個人出發，這也許和成長經驗有關，我沒有兄弟姐妹，所以從小都是自己跟自己玩。小時候玩玩具是一種對敘事的拓展，自我對話，所以我的創作基本上就是小時候玩樂高經驗的延伸。但這種個人經驗對觀眾而言一點都不重要，我是透過技術來邀請他們進入我的世界；好像送他們一個禮物，藉由精美的包裝來引誘觀眾打開，而這包裝就是技術，然後把個人故事包裝在裡面。但對觀眾而言，那個樂趣不是裡面包了什麼，而是拆封過程的樂趣。

周：你如何看待自己創作出來的角色？有時數位藝術會陷入著名的「恐怖谷效應」。那是你想透過這部作品探索的領域嗎？或是透過角色來自我挖掘？

李：我創作出的人物就是我，很難有所區隔。但我最近常被問到我是否透過創作來處理心理層面的問題或情感出口。創作的當下我沒有想那麼多，但事後仔細想想，好像有這個可能性。我是一個很有表達慾的人，我想透過替身幫我處理表達的慾望，所以是蠻輕鬆的。

周：這件作品看似有著強烈的人文主義色彩。會這麼說是因為此時此刻的數位藝術家常常陷入金融世界（加密貨幣）或某種與人工智慧的鬥爭中。然而，人工智慧在一百萬年內都永遠不可能開發出像你所創造的東西一樣奇特——我將這種差異描述為「人性」。這也許可以解釋為什麼這部作品引起許多共鳴。我想知道你是否考慮過你作品中的「奇特」與人工智慧可能出現的「怪異」之間的區別？

李：這是一個大哉問。我這次北雙的作品有一部分已經開始討論人工智慧與影像生產工具的關係，下一件作品會更深入來探討。我覺得這個問題要回到人與AI之間的區別跟定義是什麼。最近大家都在討論既然AI具有語言模型，能對談自如，但它是否真的了解語言，這是目前一個很重要辯論的重點。但另一面也反問，我們人類真的了解語言嗎？或是跟AI一樣只是在進行邏輯推演。以現在此刻來說，AI跟我之間的差別是幽默感，AI現在還做不到對於笑話的邏輯推理，這是目前我覺得人跟機器最大的差別。所以我很在乎利用幽默感來鬆動一些大的問題，這是我自己的策略。

周：如果有一天AI可以寫出笑話的話我覺得就是一件很恐怖的事。

李：或是你看到狗以帶有幽默的方式嘲笑你，人的某種特性被牠們學起來的時候，那也是挺可怕的。

周：可以透露下一件關於AI的發展的作品會是什麼？

李：我一直很好奇剛剛提到的賽伯格跟自然主義的拉扯。我認為AI是一個無法迴避的問題，我這次在北雙的作品也稍微有提到，當工具是有AI導入時，它有一個複雜的規範來限制你使用AI工具；這對我而言是一個新的概念：當工具具有使用規章。這就好比你買一個錘子，但並沒有人告訴你錘子不能用來打人，卻是用條「法律」來限制你錘子不能打人。可是目前AI的走向是，它自己本身就帶有一個限制規章。我的下部作品便是在探討人工智慧監管的問題，因為這跟我創作裡的腥羶色內容受到的規範雷同，好像我的創作自由受到了規範。但現在工具本身的自我審查，直接管制你不能生產腥羶色內容。所以我下件作品會是關注AI與人之間的關係，AI作為一個新的創作工具一方面帶來很多的可能性，但也出現更多的監管是現在看不到的。我也想從攝影史或繪畫史的角度出發，探討當時的人在面對新媒材時所出現的恐慌或規章是什麼。

周：此次展覽中是否有其他作品讓你產生強烈共鳴？

李：我很喜歡葉致甫的作品《123部奇觀預告片》，我的電影啟蒙多半來自B級片，所以他的作品讓我很有共鳴，也再次激起我對電影的熱情。我本來漸漸開始對電影感到無趣，也很久沒有密集地看電影，但他的作品讓我重新燃起想把每一部片都找出來看的慾望。某種程度上我們的出發點很接近，B級片是一種媒材感很強的影像，因為它不讓你馬上看穿故事情節，必須一直等到最後，所以很能實現電影作為一個媒材的特性。《123部奇觀預告片》很有趣，是他利用別人的素材，但很用心的思考影像的調度及音樂的配置，等於是把別人的作品重新拼湊組裝，也是很有土炮的概念。

訪談與整理：周安曼、施維麟 (William Smith)

編審：沈怡寧

本次訪談發表於小世界線上誌