대만 포기? 패배보다 더 나쁜 유화정책

[위종민]

CIMCEC에 올라온 대만 유화정책이 왜 나쁜가에 대한 장문의 글입니다. 

대만을 포기하는 대가: 유화정책이 패배보다 훨씬 더 나쁜 결과를 초래하는 이유

2026년 7월 6일

객원 기고

Josh Richards, Joseph Hanacek

패배를 감수할 가치가 있는 대의

외교 정책 논쟁은 흔히 하나의 질문을 중심으로 전개된다. 어떤 대의는 싸울 가치가 있는가? 그러나 전략가들은 훨씬 덜 자주, 더 불편하지만 궁극적으로는 더 많은 것을 드러내는 질문을 던진다. 어떤 대의는 패배를 감수할 가치가 있는가? 대만의 경우, 이 구분은 매우 중요한 의미를 갖는다.

대만을 둘러싼 전쟁은 막대한 군사적 손실, 심각한 세계 경제 충격, 그리고 핵 사용 문턱에까지 이를 수 있는 확전 양상을 초래할 가능성을 안고 있다. 미국 국방부 고위 관계자들, 동맹국 정부, 그리고 독립적인 분석가들은 점점 한 가지 점에서 의견을 같이하고 있다. 미국과 중국이 직접 충돌할 경우 확실한 것은 단 하나뿐이며, 승패와 관계없이 그 파장은 심각하고 전 세계적인 규모에 이를 것이라는 점이다.¹

그러나 전쟁의 위험성에만 초점을 맞추면 더 큰 전략적 현실을 놓치게 된다. 대만은 점점 더 복잡해지는 인도·태평양 지역의 또 하나의 분쟁 지점이 아니다. 대만은 미국의 힘이 기반을 두고 있는 현대 국제질서의 기술적·경제적·군사적 구조를 떠받치는 핵심 축이다. 중국이 대만을 강제로 자국의 통제 아래 두도록 방치하는 것은 단지 양안의 세력 균형을 바꾸는 데 그치지 않는다. 그것은 세계 권력의 분포 자체를 장기적이고도 미국과 그 동맹국들에게 매우 불리한 방향으로 재편하게 될 것이다.

미국과 대만의 관계의 핵심에는 21세기 남은 기간 전체에 영향을 미칠 요소들이 자리하고 있다. 첨단 컴퓨팅 하드웨어에 대한 접근, 미국의 외교적 신뢰성과 실질적인 전력 투사 능력, 그리고 차세대 기술 혁신을 이끌 전략적 위치가 그것이다. 대만을 포기하는 것은 이 세 가지 모두에 동시에 비용을 초래하며, 일단 그러한 비용이 발생하면 이를 되돌리는 것은 극도로 어렵거나 사실상 불가능할 것이다.²

다시 말해, 현재의 전략적 상황은 미국이 대만을 방어하다가 심각한 대가를 치를 수도 있다는 데 있는 것이 아니다. 오히려 대만을 방어하지 않기로 선택하는 것이 미국의 국력에 훨씬 더 큰 장기적 비용을 초래한다는 데 있다. 가장 바람직한 결과는 대만의 민주적 자치와 양안 현상 유지를 보존하는 평화적 해결이다. 그러나 사실상 대만을 강압적인 통일에 내맡기는 평화적 유화는 미국과 서방 세계 전체에 재앙적인 결과를 초래할 것이다.

질문의 재구성: 유화정책의 전략적 비용

질문을 올바르게 다시 설정하면, 대만 문제는 행동에 따르는 위험보다 행동하지 않았을 때의 결과에 관한 문제이다. 만약 미국이 중국의 강압이나 무력 사용 앞에서 대만을 포기한다면, 세 가지 손실이 특히 두드러질 것이며, 이는 충분히 현실적인 시나리오일 뿐 아니라 일단 현실화되면 대부분 되돌리기 어려운 성격을 지닌다.

첫 번째는 첨단 컴퓨팅 분야에서 미국의 주도권이 약화되고, 궁극적으로는 붕괴할 가능성이다. 여기에는 인공일반지능(AGI) 경쟁은 물론, 고성능 컴퓨팅과 소비자용 컴퓨팅 하드웨어 전반의 생태계가 포함된다. 대만은 단순히 세계 반도체 시장의 주요 공급국이 아니다. 대만은 세계 최첨단 반도체 제조 역량의 핵심 거점이다. 이 제조 역량을 통제하는 것은 공급망에 대한 영향력뿐 아니라 기술 발전의 속도와 방향 자체를 좌우하는 지렛대를 확보하는 것을 의미한다.³ 만약 대만의 반도체 생태계가 중화인민공화국(PRC)의 손에 넘어간다면, 중국은 인공지능, 첨단 무기체계, 그리고 광범위한 디지털 경제를 떠받치는 하드웨어 생산에서 비교할 수 없는 우위를 확보하게 될 것이다.⁴

유화정책의 두 번째 비용은 전 세계적으로 미국의 외교적 신뢰성이 심각하게 훼손되는 것이며, 특히 아시아 전역에서 미국의 전력 투사 능력에 대한 신뢰가 크게 약화되는 것이다. 대만은 제1도련선의 핵심축이며, 동맹국과 파트너 국가들이 미국의 결의를 평가하는 중심 기준이다. 미국이 대만을 방어하지 않는다면, 이는 단순히 하나의 개별 사건으로 받아들여지지 않을 것이다. 오히려 미국이 더 광범위한 안보 공약을 위해 위험을 감수할 의지가 있는지에 대한 신호로 해석될 것이다. 보다 구체적으로는, 중국은 제1도련선 전역에서 지정학적으로 막대한 영향력을 행사할 수 있는 전략적 돌출 거점을 확보하게 된다. 일본, 한국, 필리핀, 싱가포르를 비롯한 여러 국가들은 미국의 지속적인 개입 의지, 기지 사용, 정보 공유, 그리고 전략적 공조에 대해 오랫동안 유지해 온 가정을 다시 검토할 수밖에 없게 될 것이다.⁵ 역사는 후원국의 결의에 대한 신뢰보다 버림받을 것이라는 두려움이 더 커질 경우, 동맹은 빠르게 약화되며, 한 번 무너진 동맹은 이전과 같은 조건으로 거의 복원되지 않는다는 사실을 보여준다.⁶

세 번째 비용은 보다 미묘하지만 결코 덜 중대한 것이 아니다. 그것은 미국이 차세대 핵심 산업에서 선도적 위치를 확보할 수 있는 능력을 잃는 것이다. 첨단 컴퓨팅은 그 자체가 목적이 아니라, 양자기술, 자동화 제조, 로봇공학, 첨단 소재, 그리고 확장 가능한 에너지 시스템 등 미래 산업을 가능하게 하는 기반 기술이다. 미국이 현재 이들 분야에서 강력한 우위를 유지하는 이유는 자본이나 정책 때문만이 아니라, 세계 최고의 과학자와 공학 인재들에게 가장 매력적인 생태계를 여전히 갖추고 있기 때문이다.⁷ 기술적 주도권이 무너지고, 여기에 세계적 공약으로부터의 눈에 띄는 후퇴가 더해진다면, 이러한 인재의 흐름과 그들이 만들어낼 혁신은 다른 중심지로 이동하게 될 것이다.

이 세 가지 손실을 종합해 보면, 이는 단순한 지역적 후퇴를 의미하는 것이 아니다. 그것은 미국이 세계 기술 및 안보 질서의 중심적 위치를 스스로 내려놓게 되는 경로를 보여준다. 일단 이러한 과정이 시작되면, 이를 되돌리기 위해서는 막대한 자원과 시간, 그리고 정치적 결속이 필요할 것이며, 역사적으로도 전략적 주도권이 상대에게 넘어간 이후 이를 되찾기 위해 그러한 역량을 다시 결집하는 데 성공한 강대국은 드물었다.

유화정책의 첫 번째 비용: 대만과 미국 기술 패권의 기반

대만이 미국에 갖는 전략적 중요성은 흔히 한 단어로 요약된다. 바로 반도체다. 그러나 이러한 단순화는 양국 관계의 깊이를 충분히 설명하지 못하며, 대만의 역할을 대체하기가 왜 그토록 어려운지도 가려버린다. 대만은 단순히 세계 반도체 시장에서 큰 비중을 차지하는 생산국이 아니다. 대만은 세계 최첨단 반도체 제조 생태계의 핵심 거점이다. 대만 기업들은 최첨단 로직 반도체 제조, 고도화된 공정 통합, 그리고 이를 가능하게 하는 암묵적 제조 노하우를 결합하여 세계에서 가장 앞선 반도체 공정을 생산할 수 있는 유일한 업체들이다. 미국, 중국, 한국, 일본, 유럽연합이 막대한 국가 지원을 바탕으로 투자해 왔음에도, 어느 경쟁국도 대만의 최첨단 제조 역량을 의미 있는 규모로 재현하는 데 성공하지 못했다.⁸

첨단 반도체는 이제 인공지능, 첨단 무기체계, 보안 통신, 우주 및 사이버 역량, 산업 생산성을 포함한 현대 국가 권력의 거의 모든 영역을 떠받치는 기반이 되었다. 미국은 반도체 설계, 전자설계자동화(EDA), 시스템 통합 분야에서 세계 최고 수준의 경쟁력을 유지하고 있지만, 이러한 강점은 대만의 최첨단 제조 능력과 분리될 수 없다.⁹ 이러한 제조 기반에 접근하지 못한다면 미국의 기술 생태계는 취약해지고, 혁신 속도는 느려지며, 외부 충격에 훨씬 더 쉽게 흔들리게 된다.

미국이 「CHIPS and Science Act」와 같은 정책을 통해 반도체 생산을 자국으로 되돌리거나 우방국 중심으로 재편하려는 노력에 대해서는 많은 논의가 이루어져 왔다. 이러한 노력은 전략적으로 반드시 필요하지만, 그것이 곧 전략적 자급을 의미하는 것은 아니다. 몇 개의 반도체 공장을 건설한다고 해서 대만이 수십 년에 걸쳐 구축해 온 촘촘한 산업 생태계가 재현되는 것은 아니다. 그 생태계에는 전문 공급망, 숙련된 인력, 빠른 공정 개선 주기, 그리고 정치적 일정에 맞춰 법률만으로 만들어낼 수 없는 제도적·기술적 축적이 포함되어 있다.¹⁰ 가장 낙관적인 전망조차도 미국 내 반도체 공장이 충분히 성숙하기까지는 수년이 걸릴 것으로 보고 있으며, 현재 대만 기업들이 달성하고 있는 수율, 유연성, 비용 경쟁력에 근접하기까지는 그보다 더 오랜 시간이 필요할 것으로 예상된다.¹¹

전략적 관점에서 대만의 가치는 단순히 생산량에 있는 것이 아니라, 최첨단 컴퓨팅 공급망의 지속성과 신뢰성을 유지하는 데 있다. 이러한 공급망은 첨단 컴퓨팅이라 부를 수 있는 영역, 즉 인공지능 연구, 군사 시스템, 과학 연구에 사용되는 고성능 반도체는 물론, 세계 디지털 생태계를 유지하는 대중시장용 컴퓨팅 하드웨어까지 함께 뒷받침한다.

이러한 의미에서 대만은 현대 세계의 핵심 기반시설로 기능한다. 대만의 반도체 공장은 아무런 대가 없이 다른 곳으로 이전하거나 복제할 수 있는 대체 가능한 자산이 아니다. 그것들은 민주주의 체제와 권위주의 체제 사이의 기술 패권 균형을 앞으로 수십 년 동안 좌우하게 될 전략적 병목지점이다.¹²

컴퓨팅 군비 경쟁: 나쁜 시나리오, 최악의 시나리오, 그리고 되돌릴 수 없는 결과

큰 틀에서 보면, 두 가지 현실적인 미래 시나리오를 상정할 수 있다. 두 경우 모두 심각한 피해를 초래하지만, 그중 하나는 다른 하나보다 훨씬 더 불안정한 결과를 낳는다.

첫 번째는 이른바 '나쁜 시나리오(bad case)'다. 이 시나리오에서는 봉쇄, 검역, 또는 제한적인 군사 충돌로 인해 대만의 반도체 생산이 중단되거나 급격히 위축된다. 첨단 반도체 공급은 붕괴하고, 비용은 폭등하며, 인공지능 연구와 상용화의 상당 부분이 멈춰 서게 되고, 세계 경제는 급격한 침체에 빠진다. 분석가들은 대만에서 대규모 공급 차질이 발생할 경우, 첫해에만 세계 경제가 수조 달러 규모의 생산 손실을 입을 수 있다고 추정한다.¹³

이 나쁜 시나리오에서는 미국과 중국 모두 그 공백을 메우기 위해 경쟁하게 된다. 서방은 생산 역량을 재건하는 데 필요한 핵심 지식재산권, 반도체 설계 역량, 특수 제조장비의 상당 부분에 계속 접근할 수 있지만, 공장 건설, 숙련 인력 양성, 생산 확대에는 시간이 오래 걸린다. 반면 중국은 산업 동원 능력에서 강력한 우위를 갖고 있다. 즉, 신속한 건설 능력, 대규모 자본 투입, 노동력 동원, 국가 보조금을 통한 비효율성 흡수 능력을 보유하고 있다.¹⁴ 동시에 중국은 수출 통제와 제재, 그리고 최첨단 노광장비와 설계 도구에 대한 제한된 접근이라는 제약도 계속 받게 된다. 그 결과 어느 쪽도 무사히 벗어나지 못하는 길고 비용이 많이 드는 컴퓨팅 군비 경쟁이 전개될 것이다.

두 번째 시나리오인 '최악의 시나리오(worst case)'는 성격 자체가 다르다. 이 경우 대만의 반도체 생태계는 단순히 가동을 멈추는 것이 아니라, 점령, 강제 통합, 또는 미국의 사실상 포기로 인해 급속한 정치·산업적 통합이 이루어지면서 중국의 실질적인 통제 아래 들어가게 된다. 이러한 상황에서 가장 중요한 변수는 공장 시설의 피해 정도가 아니라, 대만의 인력, 지식재산권, 그리고 특수 장비가 어떤 운명을 맞느냐이다. 만약 이러한 자산들이 온전한 상태로 중국의 산업 기반에 흡수된다면, 중국은 최첨단 컴퓨팅과 소비자용 컴퓨팅 모두에서 거의 확실하게 경쟁국들을 앞서 나가게 될 것이다.¹⁵

이러한 결과는 단기적인 공급 충격을 훨씬 넘어서는 영향을 가져온다. 중국은 데이터센터용 AI 가속기와 AI 학습용 반도체에서부터 스마트폰, 전기차, 항공전자장비, 미사일 유도체계에 이르기까지 현대 컴퓨팅의 거의 모든 분야에서 첨단 하드웨어의 지배적인 생산국이 될 것이다. 인공지능 경쟁의 균형도 결정적으로 중국 쪽으로 기울게 된다. 이는 단순히 알고리즘이 더 우수해서가 아니라, 대규모 실험과 실제 배치를 가능하게 하는 핵심 하드웨어에 특권적으로 접근할 수 있기 때문이다.¹⁶ 인공일반지능(AGI)이 가까운 미래에 실현될 것이라고 보든, 먼 미래의 일이라고 보든, 혹은 그것이 인류에게 이로운 기술인지 위험한 기술인지와 관계없이, 그러한 기술의 발전이 권위주의 국가이자 미국의 전략적 경쟁국의 독점적인 통제 아래 이루어진다면, 이는 현대 역사에서 세계 권력 균형을 가장 크게 뒤흔드는 사건 가운데 하나가 될 것이다.

나쁜 시나리오만 발생하더라도 각국은 안정적인 컴퓨팅 자원 확보를 위해 경쟁하면서 세계는 서로 경쟁하는 기술 블록으로 분열되기 시작할 것이다. 그러나 최악의 시나리오에서는 이러한 분열이 빠르게 고착화되어 중국의 우위가 굳어지고, 다른 국가들은 워싱턴이 아니라 베이징이 정한 표준과 플랫폼, 그리고 의존 구조에 적응하도록 강요받게 될 것이다. 이러한 전환이 이루어진 뒤에는 이를 되돌리기 위해 단순한 정책 변화만으로는 부족하다. 산업과 연구 생태계 전체를 다시 구축해야 하며, 역사적으로 볼 때 전략적 주도권을 크게 상실한 이후 이를 성공적으로 복원한 강대국은 거의 없었다.

이러한 관점에서 보면, 대만을 포기할 경우의 가장 큰 위험은 단순한 공급 차질이 아니라 '흡수'이다. 따라서 대만의 반도체 생태계가 중국의 손에 아무런 단절 없이 넘어가는 것을 막는 일은 부차적인 우려나 비상대책이 아니라, 중국의 침략에 대응하는 미국 전략을 규정하는 핵심적인 전략적 과제 가운데 하나이다.

실리카의 평화(Pax Silica)의 종말: 두 개의 기술 생태계와 분열된 세계 경제

대만의 반도체 생태계가 붕괴하거나, 더 나아가 중국에 흡수된다면 그 결과는 단순한 반도체 부족이나 일시적인 시장 혼란에 그치지 않을 것이다. 보다 지속적인 영향은 이른바 '실리카의 평화(Pax Silica)'라 부를 수 있는 질서의 붕괴일 것이다. 이는 첨단 컴퓨팅 공급망과 기술 표준, 연구 네트워크가 정치 체제의 차이를 넘어 대체로 상호운용성을 유지하는 세계 기술 질서를 의미한다.

이러한 질서는 이미 흔들리고 있다. 미국의 수출 통제, 중국의 산업 정책, 그리고 기술 흐름을 둘러싼 전략 경쟁의 심화는 세계 기술 생태계를 점차 분열시키기 시작했다. 대만을 둘러싼 결정적인 충돌이 발생한다면 이러한 과정은 극적으로 가속화되어, 세계는 미국과 동맹국을 중심으로 한 하나의 기술 생태계와 중국을 중심으로 한 또 다른 기술 생태계라는 서로 경쟁하며 점점 더 양립하기 어려운 두 개의 기술 체계로 나뉘게 될 것이다.¹⁷

겉으로 보기에는 이러한 분열이 신흥시장에서 어떤 스마트폰이 우위를 차지하는지, 어떤 인공지능 플랫폼이 가장 널리 사용되는지, 또는 차세대 네트워크를 어떤 디지털 표준이 지배하는지와 같은 소비자 중심의 선택 문제처럼 보일 수도 있다. 그러나 실제로는 훨씬 더 깊은 차원의 균열이 발생하게 된다. 경쟁하는 기술 생태계는 희토류, 첨단 소재와 화학물질, 에너지 자원, 클라우드 인프라, 금융 서비스, 그리고 장기적인 혁신을 뒷받침하는 연구 생태계에 대한 접근을 결정하게 될 것이다.¹⁸ 각국은 암묵적으로든 명시적으로든 자국의 규제 체계, 공급망, 안보 협력 관계를 어느 한 진영에 맞추도록 압력을 받게 될 것이다.

이러한 세계에서 대만의 운명은 결정적인 의미를 갖게 된다. 대만이 중국의 통제 밖에 남아 있다면, 미국과 동맹국들은 압박 속에서도 혁신과 회복력을 유지할 수 있는 최첨단 기술 생태계의 신뢰할 만한 기반을 계속 확보하게 된다. 반대로 대만이 중국의 영향권에 들어간다면 균형은 결정적으로 중국 쪽으로 기울게 된다. 중국은 첨단 하드웨어 생산에서 지배적인 위치를 차지할 뿐 아니라, 기술 표준을 설정하고 접근 조건을 결정하며, 세계 경제의 광범위한 영역에서 형성되는 의존 관계를 지렛대로 활용할 수 있는 위치에 서게 된다.¹⁹

이러한 변화는 특히 중견국과 개발도상국에 큰 불안정을 초래할 것이다. 이들 국가 가운데 상당수는 서방의 안보 보장과 중국과의 경제 협력에 동시에 의존하고 있기 때문이다. 시간이 흐를수록 이러한 압력은 탈냉전 이후 세계 경제를 특징지어 온 개방성을 약화시키고, 시장 효율성이 아니라 어느 기술 진영에 속하느냐에 의해 규정되는 보다 경직된 양극 체제로 세계를 바꾸어 놓을 것이다.²⁰

미국에게 실리카의 평화가 끝난다는 것은 단순히 편의성이나 수익성을 잃는 문제가 아니다. 그것은 국제사회에서 권력이 축적되고 행사되는 방식 자체가 구조적으로 바뀌는 것을 의미한다. 미국은 오랫동안 기술적 우위를 바탕으로 지속적인 강압에 의존하지 않고도 영향력을 확대해 왔으며, 다른 국가들이 자발적으로 채택하는 기술 표준과 플랫폼, 제도 속에 미국의 선호를 내재화해 왔다. 그러나 세계가 경쟁하는 두 개의 기술 생태계로 분열되고, 특히 중국이 가장 핵심적인 하드웨어 공급망을 장악하게 된다면, 이러한 미국의 우위는 크게 약화될 것이다.

유화정책과 해외에서의 미국 영향력 약화

그러나 기술만으로 전략적 결과가 결정되는 것은 아니다. 대만의 중요성은 인도·태평양 지역의 군사적·정치적 지리에서 차지하는 역할로 인해 더욱 커진다. 이 지역에서는 미국의 군사 배치 자체만큼이나 미국의 의지를 어떻게 인식하느냐가 중요한 의미를 갖는다.

대만은 제1도련선의 중심에 위치하며, 중국 인민해방군이 서태평양으로 군사력을 투사하는 것을 제약하는 동맹 및 협력국 네트워크의 핵심 축 역할을 한다. 대만의 지속적인 자치는 중국의 군사 계획을 복잡하게 만들고, 미국의 작전 자유를 유지하며, 미국의 안보 공약이 여전히 신뢰할 만하다는 확신을 역내 파트너들에게 제공한다.²¹ 반대로 대만을 상실할 경우, 어떠한 외교적 선언으로도 쉽게 회복할 수 없는 연쇄적인 신뢰 위기가 발생하게 될 것이다.

외교적 신뢰성은 추상적인 개념이 아니다. 그것은 위험을 감수해야 하는 상황에서도 약속을 지키겠다는 행동을 반복적으로 보여줌으로써 구축된다. 만약 미국이 대만 방어에 실패한다면, 아시아 전역에서는 이를 단순히 대만만의 예외적인 사건으로 받아들이지 않을 것이다. 오히려 미국이 인도·태평양에서 전략적 지위를 유지하는 데 필요한 비용을 감당할 의지나 능력이 없다는 신호로 해석할 것이다. 일본은 확장억제의 실효성과 자국 남서도서 지역의 안보를 다시 검토해야 할 것이며, 한국은 독자적인 핵 능력을 추구해야 한다는 압력을 새롭게 받게 될 것이다. 동남아시아 국가들은 미국의 역할 축소를 예상하면서 기지 제공, 정보 협력, 외교적 노선을 재조정하는 등 균형 전략을 더욱 가속화할 것이다.²²

이러한 반응의 논리는 국가 행동을 움직이는 것은 공포, 명예, 그리고 이익이라는 투키디데스의 오래된 통찰로 잘 설명된다. 동맹이 유지되는 이유는 파트너 국가들이 이타적이기 때문이 아니라, 집단 안에 있을 때 자신의 이익과 명예가 밖에 있을 때보다 더 안전하다고 믿기 때문이다. 후원국의 의지에 대한 신뢰보다 버림받을 것이라는 두려움이 더 커지는 순간, 동맹은 유지되지 못하고 순응과 타협으로 바뀌게 된다. 역사적으로 이러한 전환을 겪고도 원래 모습 그대로 유지된 동맹은 거의 없었다.²³

중요한 점은 이러한 역학이 양방향으로 작용한다는 것이다. 미국이 불리한 상황에서도 대만을 방어하기 위해 신뢰할 만한 노력을 기울인다면, 이는 대만 해협을 넘어 훨씬 더 넓은 범위에서 억제력을 강화하게 된다. 이는 미국이 한계를 설정하면 그것을 넘는 데에는 반드시 대가가 따른다는 신호를 보내는 것이다. 외교가 이념적 틀보다 현실정치에 더욱 기반을 두게 되는 세계에서, 중국과의 벼랑 끝 대치를 감수할 의지를 보여주는 것은 동맹국과 협력국들에게 비용을 분담하고 위험을 감수하며, 민주주의 국가의 정권 교체에 따른 정치적 변화와 관계없이 장기 전략을 미국의 리더십과 맞추는 것이 자신들의 국익에도 부합한다는 확신을 심어주는 유일한 방법이다. 여러 분석가들이 지적했듯이, 신뢰성은 종종 완벽한 승리를 거두는 것보다 결과가 불확실한 상황에서도 흔들리지 않고 버티겠다는 의지를 보여주는 데서 형성된다.²⁴

따라서 대만은 전략적 리트머스 시험지와 같은 역할을 한다. 미국 정책결정자들이 직면한 질문은 대만을 방어하는 것이 안정을 반드시 보장하느냐가 아니다. 오히려 대만을 포기하는 것이 지난 반세기 이상 아시아에서 미국의 영향력을 떠받쳐 온 동맹 체제의 붕괴를 사실상 불가피하게 만드는 선택이 되는 것은 아닌가 하는 점이다.

지리는 여전히 중요하다: 현대의 '침몰하지 않는 항공모함'으로서의 대만

반도체와 인공지능이 전략 용어의 일부가 되기 훨씬 이전부터 대만의 중요성은 지리적 위치에서 비롯되었다. 더글러스 맥아더 장군은 대만을 유명하게도 '침몰하지 않는 항공모함'이라고 표현했는데, 이는 미국이 이곳을 전진기지로 삼아 아시아 해양 전역에 군사력을 투사하고 적국을 견제할 수 있다는 의미였다. 맥아더 시대 이후 전쟁 기술은 극적으로 변화했지만, 그 근본적인 논리는 변하지 않았다. 오히려 그 중요성은 더욱 커졌다.²⁵

자율성을 유지하며 미국과 협력 관계를 유지하는 대만은 역내 전반의 억제 구도를 형성하는 핵심 요소다. 현재 중국이 해양으로 군사력을 투사하려 한다면, 다층적인 방어망, 동맹국 간의 협조, 그리고 자신에게 불리한 조건에서 조기에 충돌이 확대될 가능성을 모두 고려해야 한다. 그러나 대만이 이러한 구도에서 사라진다면 인민해방군의 작전 환경은 훨씬 더 유리해진다. 억제력이 약화되는 이유는 미국의 군사력이 하루아침에 사라지기 때문이 아니라, 위험과 기회의 균형이 결정적으로 중국에 유리한 방향으로 이동하기 때문이다.²⁶

지정학적 관점에서 중국이 통제하는 대만은 전혀 다른 의미를 갖는다. 대만은 동북아와 동남아, 그리고 더 넓은 태평양을 연결하는 가장 중요한 해상 및 항공 교통로를 가로지르는 위치에 있다. 중국이 이 섬을 장악하게 되면 인민해방군은 제1도련선을 돌파하여 필리핀해 깊숙이까지 군사력을 투사할 수 있으며, 오랫동안 미국과 동맹국들에게 유리하게 작용해 온 방어 지형을 무너뜨릴 수 있다.²⁷ 또한 대만을 거점으로 미국 해군의 작전을 더욱 효과적으로 견제하고, 일본의 핵심 기지를 위협하며, 세계 무역의 상당 부분이 통과하는 해상교통로에 지속적인 압박을 가할 수 있게 된다.²⁸

그보다 더 중요한 것은 이러한 변화가 가져올 연쇄적인 효과이다. 중국이 대만을 통제하게 되면 일본에서 싱가포르에 이르는 제1도련선 전체 해역에 대한 중국의 영향력은 크게 강화된다. 만약 중국이 이 해역에서 상업 활동을 하는 국가들에 압력을 행사하려 한다면, 대만이라는 전략 거점은 상선 운항, 에너지 수송, 해저 기반시설에 대한 위험 평가 자체를 바꾸게 될 것이다. 그 결과 해상 보험료는 상승하고, 세계 무역은 정치적 압력에 훨씬 더 취약해질 것이다.²⁹ 해상 교역에 크게 의존하는 국제질서에서 이러한 변화의 충격은 인도·태평양을 훨씬 넘어 세계 전체로 확산될 것이다.

이러한 관점에서 보면 맥아더의 표현은 시대에 뒤떨어진 비유가 아니다. 대만은 여전히 '침몰하지 않는 항공모함'이다. 이는 단순히 활주로를 갖춘 군사기지라는 의미가 아니라, 역내 안정을 떠받치는 고정된 지리적 전략 자산이라는 뜻이다. 이러한 자산을 포기하는 것은 아무리 먼 거리에서 군사력을 투사하더라도 쉽게 상쇄할 수 없는 장기적인 군사적 불이익을 초래하게 될 것이다.

다음 산업 혁신의 기회를 잃는다는 것

대만을 포기하는 데 따른 전략적 결과는 오늘날의 기술이나 군사력 배치에만 국한되지 않는다. 흔히 충분히 주목받지 못하는 세 번째 비용은 앞으로 수십 년간 경제력과 군사력을 결정하게 될 차세대 산업 및 과학 혁신을 주도할 기회를 상실하는 데 있다.

첨단 컴퓨팅은 양자정보과학, 자동화 및 적층 제조, 로보틱스, 첨단 소재, 저비용 대규모 에너지 시스템 등 다양한 신흥 산업의 발전을 가능하게 하는 기반 기술이다. 이러한 모든 분야에서 연구 성과와 상용화의 속도는 점점 더 고성능 컴퓨팅, 대규모 데이터셋, 그리고 인공지능 기반 실험에 대한 접근성에 의해 좌우되고 있다. 이러한 요소들을 장악한 국가는 시간이 지날수록 더욱 커지는 누적적 우위를 확보하여 혁신을 가속화하는 동시에 경쟁국과의 격차를 더욱 벌릴 수 있다.³⁰

현재 미국이 이러한 분야에서 강력한 위치를 차지하고 있는 이유는 운명적으로 우월해서가 아니라, 세계 최고 수준의 인재를 끌어들이는 독보적으로 매력적인 생태계를 갖추고 있기 때문이다. 미국의 대학, 연구기관, 벤처투자 네트워크, 그리고 개방적인 과학 연구 문화는 여전히 세계에서 가장 뛰어난 공학자와 과학자들을 끌어들이고 있다.³¹

대만을 포기하는 것은 이러한 생태계를 위험에 빠뜨릴 수 있다. 미국의 기술 리더십이 눈에 띄게 약화되고, 여기에 국제 안보 공약으로부터의 후퇴까지 더해진다면, 인재와 투자 자금은 더욱 성장 가능성이 높거나 안정적이라고 여겨지는 다른 중심지로 이동하게 될 것이다. 이미 국가 주도의 전략기술 개발에 막대한 투자를 하고 있는 중국은, 특히 대만의 반도체 산업을 장악하여 첨단 컴퓨팅에 대한 지배적인 접근권까지 확보하게 된다면 이러한 변화를 적극적으로 활용할 수 있는 매우 유리한 위치에 서게 될 것이다.³² 이러한 분야에서 뒤처진다는 것은 단순히 미국의 경제 성장을 둔화시키는 것이 아니라, 기술적 우위가 줄어든 상황에서 정책결정자들의 전략적 선택지를 더욱 제한하게 될 것이다.³³

역사는 한 번 기술적 리더십을 잃으면 이를 되찾기가 매우 어렵다는 사실을 보여준다. 기술적 리더십은 투자와 정책만으로 유지되는 것이 아니다. 그것은 명성, 자신감, 그리고 혁신가들이 자신들이 미래를 만들어가고 있다는 믿음, 즉 남을 따라잡는 것이 아니라 미래를 선도하고 있다는 확신과 같은 무형의 요소에도 의존한다. 미국이 대만을 중국의 손에 넘기도록 방치한다면, 가장 중요한 시점에서 바로 이러한 무형의 자산들을 잃을 위험을 감수하게 되는 것이다.

이러한 의미에서 대만을 방어하는 것은 단순히 현상 유지를 위한 문제가 아니다. 그것은 미국이 앞으로도 전략적 우위의 최전선을 계속 주도할 수 있는 조건을 지키는 일이다. 유화정책은 안정을 가져다주는 것이 아니라, 미국의 미래를 저당 잡히는 선택이 될 것이다.

결론: 21세기의 공포, 명예, 그리고 이익

2천 년이 넘는 세월 전, 투키디데스는 국가의 의사결정을 움직이는 것은 공포, 명예, 그리고 이익이라고 말했다. 이 틀은 오늘날에도 여전히 중요한 시사점을 제공한다. 그것은 고전적 현실주의의 유물이 아니라, 국제정치에서 권력과 신뢰성, 그리고 인식은 서로 떼어놓을 수 없다는 사실을 일깨워 주기 때문이다.

대만 문제에서는 이 세 가지 요소가 모두 하나로 맞물린다. 공포는 분명하다. 중국이 첨단 컴퓨팅의 가장 핵심적인 기반을 장악하고, 서태평양의 군사적 지형을 지배하며, 다른 국가들의 기술 및 경제 참여 조건까지 결정하는 세계에 대한 두려움이다. 이익 역시 명확하다. 수십 년 동안 미국의 번영과 안보를 뒷받침해 온 기술적 리더십, 동맹의 신뢰성, 그리고 전략적 행동의 자유를 유지하는 것이다. 그리고 흔히 오해받는 명예란 추상적인 자존심이 아니라, 동맹국과 협력국이 미국의 약속에 보내는 신뢰, 그리고 미국이 지도력을 발휘할 수 있게 만드는 평판이라는 자산에 있다.³⁴

대만이 과연 싸울 가치가 있는 대상인지에 대해서는 이미 많은 논의가 이루어져 왔다. 그러나 중요한 질문이기는 해도 그것만으로는 충분하지 않다. 더 중요한 질문은 미국이 대만을 지키지 않기로 선택한 이후의 세계를 과연 감당할 준비가 되어 있는가 하는 점이다. 여러 근거를 종합해 보면, 그러한 세계는 지금보다 더 가난하고, 더 강압적이며, 더 불안정한 세계가 될 가능성이 높다. 기술 체계는 분열되고, 동맹은 약화되며, 규칙과 합의 대신 힘과 의존 관계가 국제질서를 규정하게 될 것이다.

이 글은 대만의 중요성이 상징성에 있는 것이 아니라 구조적 위치에 있다는 점을 주장해 왔다. 대만은 현재 국제질서를 떠받치는 기술 공급망, 동맹 네트워크, 그리고 군사적 지리의 핵심 축이다. 따라서 대만을 잃는 것은 단순한 지역적 후퇴가 아니라, 혁신과 안보, 그리고 세계 거버넌스 전반에 걸쳐 여러 세대 동안 영향을 미칠 체계적 충격이 될 것이다.

이러한 이유로 이 글은 결론이라기보다 출발점에 가깝다. 대만 방어는 군사 태세만으로 이해하거나 수행할 수 있는 문제가 아니다. 그것은 보다 폭넓은 여러 기반 조건에 달려 있다. 대규모 비대칭 방어를 지속할 수 있는 능력, 경제 및 에너지 분야의 강압을 견뎌낼 수 있는 역량, 핵심 원자재를 안정적으로 확보하는 능력, 그리고 대만과 동맹국들을 연결하는 물리적·디지털 기반시설을 보호하는 능력이 모두 필요하다. 이러한 각각의 영역은 그 자체로 별도의 심층적인 분석이 필요한 주제이다.

대만을 방어한다고 해서 손쉬운 승리나 위험 없는 결과가 보장되는 것은 아니다. 그러나 대만을 포기하는 것은 미국 스스로에게 전략적인 치명상을 입히는 선택이 될 것이며, 그 영향은 억제를 유지하는 데 드는 비용보다 훨씬 더 결정적으로 앞으로 한 세기 동안 미국의 국력을 좌우하게 될 것이다. 그런 의미에서 대만은 단순히 미국의 의지를 시험하는 대상이 아니다. 그것은 앞으로 국제질서의 세력 균형이 어느 방향으로 기울 것인지를 결정하는 지렛대가 될 수도 있는 존재이다.

조시 리처즈는 Pacific Peering의 최고사업책임자(CCO)이다. 그는 유엔(UN) SMART Cables 공동 태스크포스 운영위원회 위원으로 활동하며, 비즈니스 개발위원회를 맡고 있다. 또한 Truman National Security Project의 안보 펠로우, Aspen Institute의 기술정책 펠로우, AI2030의 선임 펠로우로 활동하고 있다.

조지프 하나체크는 미국 해군의 수상전 장교이다. 그는 캘리포니아주 샌디에이고에 위치한 해군 수상·기뢰전 개발센터(Naval Surface and Mine Warfighting Development Center) 산하 수상전 고등전술학교(Surface Advanced Warfighting School) 분견대에서 전술 교관(Warfare Tactics Instructor)으로 근무하고 있다. 본문에 제시된 견해와 의견은 저자 개인의 것이며, 미국 국방부, 해군성 또는 그 산하기관의 공식 입장을 반드시 반영하는 것은 아니다.


The Cost of Abandoning Taiwan: How Appeasement Leads to an Outcome Far Worse Than Defeat | Center for International Maritime Security

The Cost of Abandoning Taiwan: How Appeasement Leads to an Outcome Far Worse Than Defeat | Center for International Maritime Sec

The Cost of Abandoning Taiwan: How Appeasement Leads to an Outcome Far Worse Than Defeat

July 6, 2026 Guest Author 4 Comments

By Josh Richards and Joseph Hanacek

A Cause Worth Losing For

Foreign policy debates are often framed around a single question: what causes are worth fighting for? Far less often do strategists ask the more uncomfortable but ultimately more revealing question, what causes are worth losing for? In the case of Taiwan, that distinction matters profoundly.

A war over Taiwan would carry the potential for catastrophic military losses, a severe global economic shock, and escalatory dynamics that could touch the nuclear threshold. Senior U.S. defense officials, allied governments, and independent analysts increasingly agree on one point: the only certainty in a direct U.S.–China conflict is that win or lose, the repercussions would be drastic and global in scope.1

But focusing only on the dangers of war obscures the larger strategic reality. Taiwan is not simply another flashpoint in an increasingly crowded Indo-Pacific. It is a structural pillar of the modern international system, technological, economic, and military, on which U.S. power rests. Allowing Beijing to force Taiwan under its control would not merely alter the cross-Strait balance; it would reshape the global distribution of power in ways that are both durable and deeply unfavorable to the United States and its allies.

At the heart of the U.S. relationship with Taiwan are factors that will influence the rest of the twenty-first century: access to advanced computing hardware, the US diplomatic credibility and real power projection, and the strategic positioning that will drive the next generations of technological breakthroughs.  Abandoning Taiwan would impose costs across all three simultaneously, and once incurred, those costs would be extraordinarily difficult, if not impossible, to reverse.2

The strategic situation, in other words, is not that the United States might defend Taiwan and suffer severe consequences. It is that choosing not to defend Taiwan would impose even greater long-term costs on American power. The preferred outcome remains a peaceful resolution that maintains preservation of Taiwan’s democratic autonomy and the cross-Strait status quo. But a peaceful appeasement that effectively abandons Taiwan to coercive unification would be cataclysmic for America and the entirety of the western world.

Reframing the Question: The Strategic Costs of Appeasement

Reframed properly, the Taiwan question is less about the risks of action than the consequences of inaction. If the United States were to abandon Taiwan in the face of Chinese coercion or force, three losses stand out as both credible and, once triggered, largely irreversible.

The first is the erosion, and likely collapse, of U.S. leadership in advanced computing, including the race toward artificial general intelligence (AGI) and the broader ecosystem of high-performance and consumer computing hardware. Taiwan is not merely a major supplier within global semiconductor markets; it is the central node of the world’s most advanced fabrication capacity. Control over that capacity confers leverage not only over supply chains, but over the pace and direction of technological progress itself.3 Should Taiwan’s semiconductor ecosystem fall to the People’s Republic of China (PRC), China would gain an unparalleled advantage in the production of the hardware that underpins artificial intelligence, advanced weapons systems, and the digital economy writ large.4

The second cost of appeasement would be a profound loss of U.S. diplomatic credibility around the world, particularly with regards to its power projection capability across Asia. Taiwan is the keystone of the first island chain and a focal point for how allies and partners assess American resolve. A failure to defend Taiwan would not be interpreted in isolation; it would be read as a signal about U.S. willingness to absorb risk on behalf of its security commitments more broadly. In more concrete terms, it would result in the creation of an incredibly influential strategic salient for China, providing them geostrategic leverage throughout the entire first island chain. Japan, South Korea, the Philippines, Singapore, and others would be forced to reconsider long-standing assumptions about American staying power, basing access, intelligence sharing, and strategic alignment.5 History suggests that when fear eclipses confidence in a patron’s resolve, alliances erode quickly, and are rarely reconstituted on the same terms.6

The third cost is more subtle, but no less consequential: the loss of the United States’ ability to break out ahead in the next generation of foundational industries. Advanced computing is not an end in itself; it is the enabling substrate for future dominance in quantum technologies, automated manufacturing, robotics, advanced materials, and scalable energy systems. The United States currently enjoys a powerful advantage in these domains not only because of capital or policy, but because it remains the world’s most attractive ecosystem for top scientific and engineering talent.7 A collapse in technological leadership, paired with a visible retreat from global commitments, would redirect those talent flows, and the breakthroughs they enable toward alternative centers of gravity.

Taken together, these three losses describe more than a regional setback. They outline a pathway by which the United States would relinquish its position at the center of the global technological and security order. Once the process began, reversing it would demand resources, time, and political cohesion on a scale that great powers historically struggle to mobilize once strategic momentum turns against them.

The first cost of appeasement: Taiwan and the Foundations of American Technological Power

Taiwan’s strategic importance to the United States is often summarized in shorthand: semiconductors. But that simplification understates the depth of the relationship and obscures why Taiwan’s role is so difficult to replace. Taiwan is not merely a large producer within global chip markets; it is the central node of the world’s most advanced semiconductor manufacturing ecosystem. Its firms are the world’s only producers capable of manufacturing the world’s most advanced semiconductor nodes, combining leading-edge logic fabrication, advanced process integration, and the tacit manufacturing knowledge required. Despite massive state-backed investments by the United States, China, South Korea, Japan, and the European Union, no competitor has successfully replicated Taiwan’s frontier manufacturing capabilities at any scale.8

Advanced semiconductors now underpin nearly every domain of modern state power, including artificial intelligence, advanced weapons systems, secure communications, space and cyber capabilities, and industrial productivity. The United States retains world-class strength in chip design, electronic design automation, and systems integration, but those advantages are inseparable from Taiwan’s ability to manufacture at the cutting edge.9 Without access to that manufacturing base, the U.S. technology stack becomes brittle, slower to innovate, and more vulnerable to disruption.

Much has been made of U.S. efforts to reshore or “friend-shore” semiconductor production through initiatives such as the CHIPS and Science Act. These efforts are strategically necessary, but they should not be confused with strategic sufficiency. Building a handful of fabrication plants does not recreate the dense industrial ecosystem Taiwan has spent decades refining, an ecosystem that includes specialized suppliers, an experienced workforce, rapid iteration cycles, and institutional memory that cannot be legislated into existence on a political timeline.10 Even optimistic projections suggest that U.S.-based fabs will take years to reach maturity, and longer still to approach the yield, flexibility, and cost structures that Taiwanese firms currently achieve.11

From a strategic perspective, Taiwan’s value lies not only in output volume, but in the continuity and reliability of frontier computing supply. That supply underwrites what can be described as frontier computing, the high-performance chips used in advanced AI research, military systems, and scientific discovery, as well as consumer computing, the mass-market hardware that sustains global digital ecosystems.

In this sense, Taiwan functions as critical infrastructure for the modern world. Its fabs are not interchangeable assets that can be relocated or duplicated without consequence. They are strategic choke points whose fate will shape the balance of technological power between democratic and authoritarian systems for decades to come.12

The Computing Arms Race: Bad Cases, Worst Cases, and Irreversible Outcomes

At a high level, two plausible futures stand out, both damaging, but one far more destabilizing than the other.

The first is what might be termed the bad case. In this scenario, a Taiwan crisis, through blockade, quarantine, or limited conflict, causes Taiwanese semiconductor production to halt or sharply contract. Advanced chip supply would collapse, driving up costs, freezing portions of AI research and deployment, and triggering a sharp economic downturn. Analysts estimate that a major Taiwan disruption could cost the global economy trillions of dollars in lost output within the first year alone.13

In this bad case, both the United States and China would scramble to fill the void. The Western world retains access to much of the underlying intellectual property, design expertise, and niche manufacturing equipment required to rebuild capacity, but it moves slowly in constructing facilities, training labor forces, and scaling production. China, by contrast, possesses formidable advantages in industrial mobilization: the ability to build quickly, direct capital at scale, compel workforce participation, and absorb inefficiencies through state subsidy.14 At the same time, Beijing would continue to face constraints imposed by export controls, sanctions, and limited access to the most advanced lithography and design tools. The result would be a grinding, costly computing arms race in which neither side emerges unscathed.

The second scenario, the worst case, is qualitatively different. In this outcome, Taiwan’s semiconductor ecosystem does not merely shut down; it falls under effective PRC control, either through occupation, coerced integration, or a perceived abandonment by the United States that precipitates a rapid political and industrial merger. In such a case, the single most important variable is not physical plant damage, but the fate of Taiwan’s workforce, intellectual property, and specialized equipment. Should those assets be absorbed intact into China’s industrial base, Beijing would almost certainly surge ahead in both frontier and consumer computing.15

This outcome would have consequences far beyond short-term supply shocks. China would become the dominant producer of advanced hardware across nearly every category of modern computing, from data center accelerators and AI training chips to smartphones, electric vehicles, avionics, and missile guidance systems. The competitive balance in artificial intelligence would tilt decisively, not because of superior algorithms alone, but because of privileged access to the hardware that makes large-scale experimentation and deployment possible.16 Whether one believes artificial general intelligence is imminent or distant, beneficial or dangerous, its development under the exclusive control of an authoritarian state and great power rival would represent one of the most destabilizing and consequential shifts in the global balance of power in modern history.

Even in the bad case, the world would begin to fracture into competing technological blocs as states seek reliable access to computing resources. In the worst case, that fracture would harden rapidly, locking in Chinese advantages and forcing others to adapt to standards, platforms, and dependencies set in Beijing rather than Washington. Once that transition occurred, reversing it would require not just policy change, but the reconstruction of entire industrial and research ecosystems, a task that history suggests few great powers successfully accomplish after a major strategic reversal.

From this perspective, the central danger of abandoning Taiwan is not disruption, but capture. Preventing the seamless transfer of Taiwan’s semiconductor ecosystem into PRC hands is therefore not a secondary concern or contingency plan; it is one of the core strategic imperatives shaping the U.S. response to China’s aggression.

Pax Silica Ends: Two Tech Stacks and a Fractured Global Economy

If Taiwan’s semiconductor ecosystem is disrupted, or worse, captured, the consequences will not be confined to chip shortages or temporary market dislocations. The more enduring effect would be the collapse of what might be called pax silica: a global technological order in which advanced computing supply chains, standards, and research networks remain broadly interoperable across political systems.

That order is already under strain. U.S. export controls, Chinese industrial policy, and intensifying strategic competition over technology flows have begun to fragment the global technology landscape. A decisive break over Taiwan would accelerate this process dramatically, pushing the world toward two competing and increasingly incompatible technology stacks, one centered on the United States and its allies, the other organized around the People’s Republic of China.17

On the surface, this bifurcation might appear to concern consumer-facing choices: whose smartphones dominate emerging markets, which AI platforms are most widely adopted, or which digital standards govern next-generation networks. In reality, the fault lines would run much deeper. Competing tech stacks would structure access to rare earth minerals, advanced materials and chemicals, energy inputs, cloud infrastructure, financial services, and the research ecosystems that underpin long-term innovation.18 Countries would be pressured, implicitly or explicitly, to align their regulatory regimes, supply chains, and security partnerships with one bloc or the other.

In such a world, Taiwan’s fate becomes determinative. If Taiwan remains outside PRC control, the United States and its partners retain a credible foundation for a high-end technology ecosystem that can sustain innovation and resilience, even under stress. If Taiwan falls into Beijing’s orbit, the balance shifts decisively. China would not only command a dominant share of advanced hardware production; it would be positioned to shape standards, dictate terms of access, and leverage dependencies across a wide swath of the global economy.19

This would be particularly destabilizing for middle powers and developing states, many of which rely simultaneously on Western security guarantees and Chinese economic engagement. Over time, such pressures would erode the openness that has characterized the post–Cold War global economy, replacing it with a more rigid, bipolar system defined by technological allegiance rather than market efficiency.20

For the United States, the end of pax silica would represent more than a loss of convenience or profitability. It would mark a structural shift in how power is accumulated and exercised internationally. Technological leadership has long allowed Washington to amplify its influence without constant coercion, embedding U.S. preferences in standards, platforms, and institutions that others voluntarily adopt. A world split between rival tech stacks, especially one in which China controls the most critical hardware inputs, would sharply curtail that advantage.

Appeasement and the Erosion of US Influence Abroad

Technology alone, however, does not determine strategic outcomes. Taiwan’s importance is magnified by its role in the military and political geography of the Indo-Pacific, where perceptions of U.S. resolve are as consequential as force posture itself.

Taiwan sits at the center of the first island chain, anchoring a network of alliances and partnerships that constrain the People’s Liberation Army’s ability to project power into the Western Pacific. Its continued autonomy complicates Chinese military planning, preserves U.S. freedom of maneuver, and reassures regional partners that American commitments remain credible.21 Conversely, Taiwan’s loss would create a cascading crisis of confidence that no amount of declaratory policy could easily repair.

Diplomatic credibility is not an abstract concept. It is built through repeated demonstrations that commitments will be honored even when doing so entails risk. A failure to defend Taiwan would be interpreted throughout Asia not as a narrow exception, but as a signal that the United States is unwilling or unable to absorb the costs required to maintain its strategic position in the Indo-Pacific. Japan would be forced to reconsider the viability of extended deterrence and the security of its southwestern islands. South Korea would face renewed pressure to pursue independent nuclear capabilities. Southeast Asian states would accelerate hedging strategies, recalibrating basing access, intelligence cooperation, and diplomatic alignment in anticipation of a diminished U.S. role.22

The logic underlying these reactions is well captured by Thucydides’ enduring insight that fear, honor, and interest drive state behavior. Alliances endure not because partners are altruistic, but because they believe their interests and honor are safer within a collective framework than outside it. When fear of abandonment outweighs confidence in a patron’s resolve, alignment gives way to accommodation. History offers few examples of alliances that survived such a reversal intact.23

Importantly, this dynamic cuts both ways. A credible U.S. effort to defend Taiwan, even under adverse conditions, would reinforce deterrence well beyond the immediate theater. It would signal that when Washington draws a line in the sand, crossing it has repercussions. In a world where diplomacy is increasingly being made on the basis of realpolitik rather than ideological framework, demonstrating a willingness to partake in brinksmanship with PRC is the only way to assure allies and partners that it is in their own best interest to share burdens, accept risk, and align their long-term strategies with U.S. leadership regardless of the shifting political winds attendant with democratic nations. As several analysts have noted, credibility is often less about winning clean victories than about demonstrating a willingness to stand firm when outcomes are uncertain.24

Taiwan thus functions as a strategic litmus test. The question facing U.S. policymakers is not whether defending Taiwan guarantees stability, but whether abandoning it would all but ensure the unraveling of the alliance system that has underwritten American influence in Asia for more than half a century.

Geography Still Matters: Taiwan as the Modern Unsinkable Aircraft Carrier

Long before semiconductors or artificial intelligence entered the strategic lexicon, Taiwan’s importance was rooted in geography. General Douglas MacArthur famously described the island as an “unsinkable aircraft carrier,” a forward position from which the United States could project power and constrain adversaries across maritime Asia. While the technologies of warfare have changed dramatically since MacArthur’s era, the underlying logic has not; if anything, it has intensified.25

An autonomous and U.S. aligned Taiwan shapes deterrence dynamics throughout the region. As things stand today, any Chinese attempt to project power seaward must contend with layered defenses, allied coordination, and the prospect of early escalation under unfavorable conditions. Remove Taiwan from that equation, and the PLA’s operating environment becomes markedly more permissive. Deterrence weakens not because U.S. capabilities vanish overnight, but because the balance of risk and opportunity shifts decisively in Beijing’s favor.26

A PRC controlled Taiwan presents a stark contrast from a geostrategic perspective. As Taiwan sits astride the most critical maritime and air corridors linking Northeast Asia to Southeast Asia and the broader Pacific, control of the island would allow the PLA to break through the first island chain, extending Chinese power projection deep into the Philippine Sea and undermining the defensive geometry that has long favored the United States and its allies.27 From Taiwan, China could more effectively contest U.S. naval operations, threaten key bases in Japan, and exert sustained pressure on sea lines of communication that carry a significant share of global trade.28

Equally important are the second-order effects. A PRC-controlled Taiwan would significantly tighten Beijing’s grip over waters within the entire first island chain, from Japan to Singapore. In the event that China wanted to exert pressure on any nation attempting to conduct commerce inside that arc, a PRC bastion in Taiwan would alter risk calculations for commercial shipping, energy flows, and undersea infrastructure, driving up insurance costs and increasing the vulnerability of global trade to political pressure.29 For an international system heavily dependent on maritime commerce, such a shift would reverberate far beyond the Indo-Pacific.

Seen through this lens, MacArthur’s aphorism is not an anachronism. Taiwan remains an unsinkable aircraft carrier, not in the narrow sense of hosting runways, but as a fixed geographic asset that anchors regional stability. Surrendering that asset would impose enduring military disadvantages that no amount of distant force projection could easily offset.

Losing the Next Industrial Breakout

The strategic consequences of abandoning Taiwan would not end with today’s technologies or force postures. A third, often under-appreciated cost lies in the loss of opportunity to lead the next wave of industrial and scientific breakthroughs, those that will define economic and military power in the decades ahead.

Advanced computing is the enabling substrate for progress across a wide range of emerging sectors, including quantum information science, automated and additive manufacturing, robotics, advanced materials, and scalable low-cost energy systems. In each of these fields, the pace of discovery and commercialization is increasingly shaped by access to high-performance computing, large datasets, and AI-driven experimentation. States that command these inputs gain a compounding advantage, accelerating innovation while widening the gap with competitors.30

The United States currently occupies a strong position in this landscape not because of deterministic superiority, but because it hosts a uniquely attractive ecosystem for global talent. Its universities, research institutions, venture capital networks, and open scientific culture continue to draw many of the world’s most capable engineers and scientists.31

Abandoning Taiwan would place this ecosystem at risk. A visible collapse in U.S. technological leadership, paired with a retreat from global security commitments, would redirect talent flows and investment decisions toward alternative hubs perceived as ascendant or more stable. China, already investing heavily in strategic technologies through state-directed programs, would be well positioned to capitalize on such a shift, particularly if it also secured dominant access to advanced compute through control of Taiwan’s semiconductor industry.32 Losing ground in these areas would not simply slow U.S. growth; it would constrain strategic choice, forcing policymakers to operate within narrower margins of technological advantage.33

History suggests that technological leadership, once ceded, is extraordinarily difficult to reclaim. It depends not only on investment and policy, but on intangible factors, prestige, confidence, and the belief among innovators that they are building the future rather than catching up to it. By allowing Taiwan to fall into Beijing’s hands, the United States would risk forfeiting precisely those intangibles at a moment when they matter most.

In this sense, the defense of Taiwan is not merely about preserving the status quo. It is about protecting the conditions under which the United States can continue to shape the frontier of American strategic advantage. Appeasement would not buy stability; it would mortgage the future.

Conclusion: Fear, Honor, and Interest in the Twenty-First Century

More than two millennia ago, Thucydides observed that fear, honor, and interest drive the decisions of states. That framework remains instructive today, not as a relic of classical realism, but as a reminder that power, credibility, and perception are inseparable in international politics.

In the case of Taiwan, all three align. The fear is evident: a world in which China controls the most critical nodes of advanced computing, dominates the military geography of the Western Pacific, and sets the terms of technological and economic participation for others. The interest is unmistakable: preserving the foundations of American technological leadership, alliance credibility, and freedom of maneuver in a system that has underwritten U.S. prosperity and security for decades. And honor, often misunderstood, lies not in abstract notions of pride, but in the trust that allies and partners place in American commitments, and in the reputational capital that makes leadership possible.34

Much has been written about whether Taiwan is a cause worth fighting for. That question, while important, is incomplete. The more consequential question is whether the United States is prepared to live with the world that would follow from choosing not to. The evidence suggests that such a world would be poorer, more coercive, and less stable, defined by fractured technology systems, weakened alliances, and a strategic environment in which force and dependency replace rules and consent.

This paper has argued that Taiwan’s significance lies not in symbolism, but in structure. Taiwan anchors the technological supply chains, alliance networks, and military geography that sustain the current international order. Its loss would not be a discrete regional setback, but a systemic shock whose effects would reverberate across innovation, security, and global governance for generations.

For that reason, this paper serves as a foundation rather than a conclusion. The defense of Taiwan cannot be understood, or executed, through military posture alone. It depends on a wider set of enabling conditions: the ability to sustain asymmetric defense at scale, to endure economic and energy coercion, to secure critical materials, and to preserve the physical and digital infrastructure that connects Taiwan to its allies. Each of these domains merits focused analysis in its own right.

Defending Taiwan does not promise easy victories or risk-free outcomes. But abandoning it would represent a strategic self-inflicted wound, one that would shape the next century of American power far more decisively than the costs of deterrence ever could. In that sense, Taiwan is not merely a test of resolve. It is the fulcrum on which the future balance of the international system may well turn.

Josh Richards is the Chief Commercial Officer of Pacific Peering. He serves on the UN’s Joint Task Force on SMART Cables as a member of the Steering Committee, and chairs the Business Development Committee. He is a Security Fellow with the Truman National Security Project, a Tech Policy Fellow with the Aspen Institute, and a Senior Fellow with AI2030.

Joseph Hanacek is a Surface Warfare Officer in the United States Navy. He serves as a Warfare Tactics Instructor at the Surface Advanced Warfighting School detachment of the Naval Surface and Mine Warfighting Development Center in San Diego, CA. The views and opinions presented herein are those of the author and do not necessarily represent the views of the Department of War, the Department of the Navy, or its components.

References

1. U.S. Department of Defense, Military and Security Developments Involving the People’s Republic of China 2023, Office of the Secretary of Defense (Washington, DC, 2023), https://media.defense.gov/2023/Oct/19/2003323409/-1/-1/1/2023-MILITARY-AND-SECURITY-DEVELOPMENTS-INVOLVING-THE-PEOPLES-REPUBLIC-OF-CHINA.PDF

2. Hal Brands and Michael Beckley, Danger Zone: The Coming Conflict with China (New York: W. W. Norton, 2022), https://wwnorton.com/books/9781324021308

3. Chris Miller, Chip War: The Fight for the World’s Most Critical Technology (New York: Scribner, 2022), https://www.simonandschuster.com/books/Chip-War/Chris-Miller/9781982172008

4. Emily de La Bruyère, “Made in China 2025—Who Is Winning?,” Congressional Testimony: Foundation for Defense of Democracies, U.S.-China Economic and Security Review Commission February 6, 2025, https://www.uscc.gov/sites/default/files/2025-02/Emily_de_La_Bruyere_Testimony.pdf

5. Ryan Hass, “An American perspective on the role of Taiwan in US-China relations,” Brookings Institution, July, 2022, https://www.brookings.edu/articles/an-american-perspective-on-the-role-of-taiwan-in-us-china-relations/

6. Thucydides, History of the Peloponnesian War, trans. Thomas Hobbes (London: 1629), https://archive.org/details/eightbookesofpel00thucuoft/page/n5/mode/2up; see also Graham Allison, Destined for War (Boston: Houghton Mifflin Harcourt, 2017), https://www.jstor.org/stable/26557386

7.  Michael Beckley, “The Power of Nations: Measuring What Matters,” International Security 43, no. 2 (Fall 2018): 7–44, https://www.mitpressjournals.org/doi/10.1162/isec_a_00328

8. Chris Miller, Chip War: The Fight for the World’s Most Critical Technology (New York: Scribner, 2022), https://www.simonandschuster.com/books/Chip-War/Chris-Miller/9781982172008

9. Dan Blumenthal et al., “How China Views It: Sino-American Technology Competition,” American Enterprise Institute, October 2022, https://www.aei.org/research-products/report/how-china-views-it-sino-american-technology-competition/

10. Semiconductor Industry Association, “2023 State of the U.S. Semiconductor Industry,” May 2023, https://www.semiconductors.org/state-of-the-u-s-semiconductor-industry/

11. U.S. Government Accountability Office, “Semiconductor Supply Chain: Policy Considerations from Selected Experts for Reducing Risks and Mitigating Shortages,” GAO-22-105923, July 2022, https://www.gao.gov/assets/gao-22-105923.pdf

12. Emily de La Bruyère, “Made in China 2025—Who Is Winning?,” Congressional Testimony: Foundation for Defense of Democracies, U.S.-China Economic and Security Review Commission February 6, 2025, https://www.uscc.gov/sites/default/files/2025-02/Emily_de_La_Bruyere_Testimony.pdf ; see also Richard Danzig and Lorand Laskai, “Symbiosis and Strive: Where Is the Sino-American Relationship Bound?,” Johns Hopkins University Applied Physics Laboratory (JHU APL), 2020, https://www.jhuapl.edu/assessing-us-china-technology-connections/dist/00f3f3c246ab508f9fe11452bb18200c.pdf

13. Bloomberg Economics, “Xi, Biden and the $10 Trillion Cost of War Over Taiwan,” January 8, 2024, https://www.bloomberg.com/news/features/2024-01-09/if-china-invades-taiwan-it-would-cost-world-economy-10-trillion

14. Michael Beckley and Hal Brands, “The End of China’s Rise” Foreign Affairs (October 1, 2021), https://www.foreignaffairs.com/articles/china/2021-10-01/end-chinas-rise

15. Rush Doshi, Emily de La Bruyere, Nathan Picarsic, and John Ferguson, “China as a ‘Cyber Great Power:’ Beijing’s Two Voices in Telecommunications,” The Brookings Institution, April 2021 https://www.brookings.edu/articles/china-as-a-cyber-great-power-beijings-two-voices-in-telecommunications/

16. James Manyika et al., “The State of AI in 2023,” McKinsey Global Institute, August 2023, https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-in-2023-generative-ais-breakout-year

17. Emily de La Bruyère and Nathan Picarsic, “Commanding Heights: Ensuring U.S. Leadership in the Critical and Emerging Technologies of the 21st Century,” Foundation for Defense of Democracies, July 26, 2023, https://www.fdd.org/analysis/testimonies/2023/07/26/commanding-heights-ensuring-u-s-leadership-in-the-critical-and-emerging-technologies-of-the-21st-century/

18. Elizabeth Economy, The World According to China (Cambridge: Polity Press, 2021), https://www.cfr.org/book/world-according-china

19. Jude Blanchette, “China’s New Red Guards,” Foreign Affairs, June 2019, https://www.foreignaffairs.com/reviews/capsule-review/2019-10-15/two-books-china

20. Minxin Pei, “China’s Coming Upheaval,” Foreign Affairs, April 2020, https://www.foreignaffairs.com/articles/united-states/2020-04-03/chinas-coming-upheaval

21. Ryan Hass and Bonnie Glaser, “U.S.-Taiwan Relations: Will China’s Challenge Lead to a Crisis?,” Brookings Institution, 2023, https://www.jstor.org/stable/10.7864/jj.3919358

22. Zack Cooper, “Rethinking the Rebalance,” American Enterprise Institute, September 2025, https://www.aei.org/wp-content/uploads/2025/09/Rethinking-the-Rebalance-Zack-Cooper-9-5-2025.pdf?x85095

23. Thucydides, History of the Peloponnesian War, trans. Thomas Hobbes (London, 1629); see also Charles Glaser, “Rational Theory of International Politics,” International Security 41, no. 4 (Spring 2017): 49–86, https://www.mitpressjournals.org/doi/10.1162/ISEC_a_00273

24. Evan S. Medeiros, “The Changing Fundamentals of U.S.–China Relations,” The Washington Quarterly 42, no. 3 (2019): 93–119, https://www.tandfonline.com/doi/full/10.1080/0163660X.2019.1666355

25. Bonny Lin et al., Regional Responses to U.S.–China Competition in the Indo-Pacific, RAND Corporation, 2020, https://www.rand.org/pubs/research_reports/RR4412.html

26. Sidharth Kaushal, “All Strategies Short of War,” Royal United Services Institute, February 2022, https://mwi.usma.edu/all-strategies-short-of-war-getting-the-most-out-of-the-gray-zone/

27. Andrew S. Erickson and Gabriel Collins, “China’s Real Blue Water Navy,” Foreign Affairs, November/December 2012, https://www.andrewerickson.com/2012/08/chinas-real-blue-water-navy/

28. James Holmes, “Taiwan Must Prepare for War with China,” The National Interest, June 5, 2024, https://nationalinterest.org/blog/buzz/taiwan-must-prepare-war-china-211304

29. Isaac Kardon and Wendy Leutert, “Pier Competitor: China’s Power Position in Global Ports,” International Security 46, no. 4 (Spring 2022): 9–47, https://www.mitpressjournals.org/doi/10.1162/isec_a_00433

30. Michael Horowitz et al., The Future of Military Applications of Artificial Intelligence, Orbits Volume 64, Issue 4, 2020, Pages 528-543, August 2020, https://www.sciencedirect.com/science/article/abs/pii/S0030438720300430?via%3Dihub

31. Michael Beckley, “The Power of Nations: Measuring What Matters,” International Security 43, no. 2 (Fall 2018): 7–44, https://www.mitpressjournals.org/doi/10.1162/isec_a_00328

32. Rush Doshi, The Long Game: China’s Grand Strategy to Displace American Order (Oxford: Oxford University Press, 2021), https://www.rushdoshi.com/thelonggame

33. Kori Schake, “Ending China’s Chokehold on Rare-Earth Minerals,” Bloomberg, September 2020, https://www.bloomberg.com/opinion/articles/2020-09-18/ending-china-s-chokehold-on-rare-earth-minerals

34. Thucydides, History of the Peloponnesian War, trans. Thomas Hobbes (London: 1629), https://archive.org/details/eightbookesofpel00thucuoft/page/n5/mode/2up; see also Kori Schake, “Deterrence,” Hoover Institution, May 2022, https://www.hoover.org/research/deterrence

Featured image: Taipei skyline at night. (Wikimedia Commons)