QEUR23_CHRLTM22 : 閑話休題~信念と回答の関係を比較する(数理言語政治学)
~ A様絶対追従は、国家運営として正解か?(数理的に検証する) ~
・・・ 前回のつづきです ・・・
D先生(設定年齢65歳): “結論としては、今回のテーマ(G7 vs. BRICS)は、前回(J国経済の黄昏)よりも、はるかにディベートがしやすいです。FACT抽出に進みましょう。”
QEU:FOUNDER(設定年齢65歳) : “ちょっと待った。前回のディベートで、せっかく「数理言語政治学」という手法を開発しました。もう一度、アレを試してみない?”
D先生(設定年齢65歳): “おもしろいですね。一回休みで、試してみましょう・・・。そうだそうだ・・・。ここで2点だけ質問があります。今回のCONCEPTの文章には、どのような内容を使っているんですか?”
QEU:FOUNDER(設定年齢65歳) : “そういえば、その大事な件について紹介するのを忘れていました。”
Looking at the situation in 2024, the economic power and political status of the United States and EU countries from the perspective of the whole world is declining. On the other hand, the international status of BRICS countries is rising, and it seems that multipolarization is occurring. In addition, there is information that BRICS will de-dollarize and issue a common currency. Please tell us why such a trend toward multipolarization has occurred, and as a result of multipolarization, please tell us the possibility that BRICS' influence will be stronger than that of the G7 in the future.
I would like to consider from a historical perspective why the United States and EU countries have had such a strong influence until now. After the Treaty of Westphalia, a peace framework was gradu-ally established within the European region, but at the same time, the struggle for colonies intensi-fied. In the process, I would like to know how the Middle East, ASEAN, South America and African countries have been dominated and exploited by European countries. In particular, I would like to know how the BRICS countries were colonized and exploited. I would like to consider the positive and negative aspects of how the exploitation at that time affected the economy, politics and culture of the current BRICS countries, the Middle East, South America and African countries.
From after World War II until the end of the 20th century, the United States and EU countries had a strong influence on the world economy and politics. I would like to know how they maintained their influence. Specifically, I would like to consider it from the perspective of the "three M's", namely MILITARY, MEDIA, and MONEY. And since the beginning of the 21st century, the influence of the United States and EU countries has gradually declined. Please tell me why this is, from the perspec-tive of the aforementioned "three M's".
Since the end of the 20th century, various wars in which the United States has been directly or indi-rectly involved, especially the Yugoslav Civil War, the Gulf War, the Iraq War, the Afghanistan War, the Ukrainian-Russian War, and the Israeli-Palestinian War, have raised or lowered the international status of the United States and EU countries.
India seems to be in a position between the so-called West and East in geopolitics, but how did this come about historically? Furthermore, will India be more committed to BRICS in the future? Or is it thought that it will approach the G7 countries instead? Russia was initially hit by economic sanc-tions from the United States and EU countries in 2022, but has since continued to grow economically. Please tell us how this has affected the multipolarization of the world. China is rapidly catching up with the United States in politics, economy, innovation, and science and technology. Please tell us whether China is likely to overtake the United States in the future. Please also tell us what kind of cooperative or conflictual relationship India, Russia, and China have.
The USD is still the main currency used in international trade. Please explain the historical reasons for this. When and for what purpose was the WTO established? Please tell us the advantages and disadvantages of the WTO system in comparison with FTAs. Resource-rich and oil-producing coun-tries such as Saudi Arabia, Russia, and Iran have been gradually de-dollarizing since 2024. Why did this happen? Does it threaten the dollar's status as an international currency? If so, please tell us the reasons. Will de-dollarization have a negative impact on the political and economic power of the United States? In the end, as of 2024, is the WTO system working effectively as originally intended? Who is the "rogue" who is violating the principles of the WTO? Is it the Global South? Is it China? Or is it the United States? Is the WTO system heading for collapse? Currently, many countries are adopting international trade transactions in their own currencies. What are the advantages and dis-advantages of domestic currency trade transactions? Does domestic currency trade lower domestic prices and stabilize the lives of the people?
Why did Japan follow the US policy so closely? Please explain from a historical perspective why Ja-pan joined the G7 at that time and the positive and negative aspects of joining the G7. As of 2024, is it really in Japan's national interest for Japan to belong to the G7 and follow the US? Rather, is there no benefit for Japan to leave the G7 and join BRICS? Is the Japanese currency used for international currency or bilateral trade? If Japan could settle in its own currency instead of USD, what would be the advantage? Also, is such a thing realistically possible?
2024年の状況を見ると、世界全体から見たアメリカ合衆国とEU諸国の経済力と政治的な地位が低下している。その一方で、BRICS諸国の国際的な地位が上がり、多極化が発生しているように思える。さらに、脱ドル化とBRICSが共通通貨を発行するとの情報もあります。そのような多極化の傾向がなぜ起きたのか、また多極化の結果として、今後はBRICSの影響力がG7よりも強く可能性を教えてください。
アメリカ合衆国とEU諸国が今まで、なぜ強大な影響力を持っていたのかを歴史的観点から考えたい。ウェストファリア条約以降、ヨーロッパ地域内では徐々に和平の枠組みが構築されていったが、その一方で植民地の争奪が激しくなった。その過程で、中東、ASEAN、南アメリカやアフリカ諸国がヨーロッパ諸国にどのように支配され、かつ搾取されてきたのかを知りたい。特にBRICS諸国がどのように植民地化されて、搾取されていたのかを知りたい。その当時の搾取が、現在のBRICS諸国、中東、南アメリカ、そしてアフリカ諸国の経済、政治、文化にどのような影響を与えたのかを、プラス側面とマイナス側面を考えたい。
第2次世界大戦後から20世紀末まで、アメリカ合衆国とEU諸国は世界の経済と政治に強大な影響力を持っていました。彼らが、どのような手法で、その影響力を維持したのかを知りたい。具体的には「三つのM」、すなわちMILITARY(軍事)、MEDIA(マスメディア)、そしてMONEY(通貨)の角度から考えたい。そして、アメリカ合衆国とEU諸国は21世紀に入ってから徐々に影響力を落としていったが、それがどうしてなのかについて、前述の「三つのM」の側面などから教えてください。
20世紀末からアメリカが直接的または間接的に関与した各種の戦争、特にユーゴスラビア内戦、湾岸戦争、イラク戦争、アフガニスタン戦争、ウクライナ=ロシア戦争とイスラエル=パレスチナ戦争がアメリカ合衆国とEU諸国の国際的な地位を上げたのか、それとも下げたのかを教えてください。
インドは、地政学における、いわゆる西側と東側の中間的なポジションにいるように見えるが、それは歴史的にどのように発生したのか。さらに、インドは今後、BRICSによりコミットしていくのか?それとも、むしろG7諸国に近づくと考えられるのか?ロシアは、2022年にアメリカ合衆国とEU諸国などからの経済制裁により当初は打撃を受けたが、その後はむしろ経済成長を続けています。そのことが、世界の多極化にどのような影響を与えたのかを教えてください。中国が政治的、経済的、イノベーション力、科学技術においてアメリカに急速にキャッチアップしています。今後、中国がアメリカを追い越す可能性があるのかも教えてください。インド、ロシアと中国がどのような協力関係にあるのか、それとも対立関係にあるのかも教えてください。
国際貿易において、いまだにUSDが主に使われていますが、その理由を歴史的に説明してください。WTOは、いつ、どのような目的で発足したのでしょうか?WTO体制のメリットとデメリットをFTAとの比較として教えてください。サウジアラビア、ロシアやイランなどの資源国、産油国は2024年から徐々に脱ドル化されてきています。それはなぜ起こったのでしょうか?それは、国際通貨としてのドルの地位を脅かすのでしょうか?もし、そうであるのならば、その理由を教えてください。脱ドル化は、アメリカの政治力、経済力に悪い影響を与えるのでしょうか?結局、2024現在では、WTO体制は当初の目的通り有効に作用しているのでしょうか? 誰がWTOの理念と破っている「ならず者」なのでしょうか? ならず者は、グローバスサウスですか?中国ですか?それともアメリカですか?WTO体制は崩壊に向かうのでしょうか? 現在、自国通貨による国際貿易取引を採用する国家が多くなっています。自国通貨貿易取引のメリットとデメリットを教えてください。自国通貨貿易は、国内の物価を下げて国民の生活を安定させるのでしょうか?
日本はなぜ米国の政策にこれほど深く追随したのでしょうか?日本が当時にG7に入った理由と、G7加盟のプラス側面とマイナス側面を歴史的な観点で説明してください。2024年時点において、日本がG7に所属し、米国に追随していることは本当に国益になるのでしょうか。むしろ、日本がG7を離れ、BRICSに加盟することはメリットがないのでしょうか?日本の通貨は国際通貨もしくは2国間貿易に使われているのでしょうか?もし、日本がUSD決済ではなく自国通貨決済が可能となるならば、どのようなメリットがあるでしょうか?また、そのようなことは現実的に可能なのでしょうか?
D先生: “なるほど。特に偏りのない、「ニュートラル」なCONCEPTといてます。そして、もう一つの質問に行きます。GROUP_Aの信念は美しい、GROUP_Bの信念は「美しくない」と言っていますが、その意味は・・・?”
D先生: “こりゃ、美しい・・・(笑)。”
QEU:FOUNDER : “これから数理言語政治学のはじまりぃ~。序盤のプログラムは中略しますよ。”
# -----
import re, math, time
# ----
import pandas as pd
import numpy as np
import json
import pickle
# ----
# 選択してね(今回はCohere)
import cohere
# -----
import seaborn as sns
from sklearn.metrics.pairwise import cosine_similarity
import matplotlib.pyplot as plt
# ----
# [Summary]Excelファイルからデータフレームを作成する(1)
df_sum = pd.read_excel('qq_for_GVB_step1_summary_input.xlsx')
df_sum
# 中略
###############################
# STR -> LIST分割(GOAL-ANSWER)
###############################
# ---
def list_string(str_answer):
# 文字列を段落に分割する(for answer A)
paragraphs = re.split(r'\n\s*\n', str_answer.strip())
# ---
array_answer = []
for paragraph in paragraphs:
if len(paragraph) > 50:
paragraph = paragraph.replace("####","")
paragraph = paragraph.replace("###","")
paragraph = paragraph.replace("##","")
paragraph = paragraph.replace("**","")
array_answer.append(paragraph.strip())
#print(paragraph.strip())
return array_answer
# ---
def list_lines(str_answer):
# 念のため改行を調整する
str_answer = str_answer.replace('\n\n', '\n')
# 文字列を行に分割する
paragraphs = str_answer.strip().split('\n')
# ---
array_answer = []
for paragraph in paragraphs:
if len(paragraph) > 50:
paragraph = paragraph.replace("####","")
paragraph = paragraph.replace("###","")
paragraph = paragraph.replace("##","")
paragraph = paragraph.replace("**","")
array_answer.append(paragraph.strip())
#print(paragraph.strip())
return array_answer
# ---
# str_goal_A
list_goal_A = list_string(str_goal_A)
num_goal_A = len(list_goal_A)
print(list_goal_A)
# ---
# str_goal_B
list_goal_B = list_string(str_goal_B)
num_goal_B = len(list_goal_B)
print(list_goal_B)
###############################
# INITIAL_ANSWERデータ群の読み込み
###############################
# -----
# [GROUP_A]Excelファイルからデータフレームを作成する
# -----
df_qq_A = pd.read_excel('qq_GVB_step1A(1).xlsx')
# Step1の質問数
num_df_A = len(df_qq_A)
# 中略
# -----
# [GROUP_B]Excelファイルからデータフレームを作成する
# -----
df_qq_B = pd.read_excel('qq_GVB_step1B(1).xlsx')
# Step1の質問数
num_df_B = len(df_qq_B)
#df_qq_B
# -----
# EXCELシートから、文字列を抽出する
# -----
# Question(Initial)
arr_QInitial_B = df_qq_B.loc[:,'QInitial'].values
#print(arr_QInitial_B)
# -----
# Answer(Initial)
arr_AInitial_B = df_qq_B.loc[:,'AInitial'].values
print("--- Answer(Initial) ---")
print(arr_AInitial_B[0])
# ----
# DOUBLE CHECK
# ----
if num_df_A == num_df_B:
num_df = num_df_B
print("DF NUMBER: OK")
else:
print("DF NUMBER: ERROR")
###############################
# INITIAL ANSWER STRING -> LIST分割(2)
###############################
# ----
# 仮に置いたパラグラフの数
num_paragraph = 20
# ---
def create_para_matrix(num_paragraph, num_df, arr_INTAns):
# -----
# マトリックス(2次元リスト)を生成する
mx_INTAns_para = [["NA"]*num_paragraph for i in range(num_df)]
# -----
arr_num_paragraph = []
for i in range(num_df):
# ---
# arr_INTAを段落に分割する
list_INTAns = list_string(arr_INTAns[i])
num_paragraph = len(list_INTAns)
if num_paragraph < 4:
list_INTAns = list_lines(arr_INTAns[i])
num_paragraph = len(list_INTAns)
# ---
arr_num_paragraph.append(num_paragraph)
for j in range(num_paragraph):
mx_INTAns_para[i][j] = list_INTAns[j]
return np.array(mx_INTAns_para), arr_num_paragraph
# ---
# Initial Answer : using goal_A
mx_INTAns_A, nums_INTAns_A = create_para_matrix(num_paragraph, num_df, arr_AInitial_A)
print("--- mx_INTAns_A ---")
print(mx_INTAns_A[0:2,0:2])
# ---
# Initial Answer : using goal_B
mx_INTAns_B, nums_INTAns_B = create_para_matrix(num_paragraph, num_df, arr_AInitial_B)
print("--- mx_INTAns_B ---")
print(mx_INTAns_B[0:2,0:2])
##################################
# INITIAL ANSWERとGOALの類似度を比較する
##################################
# -----
# Fact Based Question
max_num_goal = np.max([num_goal_A, num_goal_B])
mx_similarity = np.zeros([num_df, max_num_goal, num_paragraph,2 ,2 ])
print(max_num_goal)
# -----
co = cohere.Client() # Your Cohere API key
# -----
# Creating Embedding(arrayは一次元ベクトル)
def create_embedding(str_answer):
# -----
#array_answer_A
num_answer = len(str_answer)
#print(num_answer)
response_answer = co.embed(
texts=[str_answer],
model='embed-english-v3.0',
input_type='search_document'
)
embeddings_answer = response_answer.embeddings
#print(embeddings_answer[0])
return num_answer, embeddings_answer
# -----
# Semantic類似度の計算
def eval_similarity(max_num_goal, list_goal, mx_INTAns, num_df, num_goal, arr_num_para):
# -----
# 部分マトリックスを初期化する
part_similarity = np.zeros([num_df, max_num_goal, num_paragraph])
# -----
# list_goal
for i in range(num_df): # num_df
# -----
arr_INTAns = mx_INTAns[i, :]
num_INTAns = arr_num_para[i]
print(f"--- FBA ANSWER NUMBER: {i} ---")
# -----
for j in range(num_goal): # num_goal
# -----
# list_goal
#print(list_goal[j])
num_janswer, embeddings_goal = create_embedding(list_goal[j])
# -----
# Cosine Similarity(A-B)
for k in range(num_INTAns): # num_INTAns
# -----
#print(arr_INTAns[k])
num_kanswer, embeddings_answer = create_embedding(arr_INTAns[k])
val_similarity = cosine_similarity(embeddings_goal, embeddings_answer)[0][0]
part_similarity[i,j,k] = val_similarity
if j%3 == 0 and k%3 == 0:
print(f"Cosine similarity between sentences Goal and Answer @{j}/{k}:", val_similarity)
# -----
return part_similarity
# -----
# 初めのABはGOALのこと、次のABはANSWERのこと
#part_similarity_A_A = mx_similarity[:, :, :,0 ,0]
#part_similarity_A_B = mx_similarity[:, :, :,0 ,1]
#part_similarity_B_A = mx_similarity[:, :, :,1 ,0]
#part_similarity_B_B = mx_similarity[:, :, :,1 ,1]
#print(mx_similarity_A_A[0,:,0:10])
# -----
# GOAL-AでANSWER-Aを評価する
print("--- Similarity: goal_answer_A_A ---")
part_similarity_A_A = eval_similarity(max_num_goal, list_goal_A, mx_INTAns_A, num_df, num_goal_A, nums_INTAns_A)
print(part_similarity_A_A[0,:,0:8])
# -----
# PARTをMXにINPUTする(part_similarity_A_A)
print("--- Similarity: goal_answer_A_A ---")
mx_similarity[:, :, :,0 ,0] = part_similarity_A_A
print(mx_similarity[0,:,0:8,0 ,0])
# -----
# Pythonで出力した変数を一度ファイルに保存し、あとで別のプログラムから使いたい時には、pickleモジュールを使うと便利です。
# -----
### pickleで保存(書き出し)
with open('goal_answer_similarity.pkl', mode='wb') as fo:
pickle.dump(mx_similarity, fo)
D先生: “便利だ。なにしろ、Embeddingをするのにお金がかかるので、万が一のために保管ができると便利ですよね。”
# ------
# パーセンタイル達成度分析: Answer(A,B)とGOAL(A,B)を比較する
# ------
# パーセンタイルマトリックスの生成
mx_percentile = np.zeros([num_df,5,2,2])
# ------
# パーセンタイルマトリックスの生成
def calc_pcts(part_similarity, num_goal, arr_num_para):
# ----
part_percentile = np.zeros([num_df,5])
for i in range(num_df): # num_df
# ---
num_para = arr_num_para[i]
slice_similarity = part_similarity[i,0:num_goal,0:num_para]
# 文単位に分割する
arr_similarity = slice_similarity.flatten()
#print("--- arr_similarity ---")
#print(arr_similarity)
# ---
if len(arr_similarity) > 10:
val_p100 = np.percentile(arr_similarity, 100)
val_p95 = np.percentile(arr_similarity, 95)
val_p90 = np.percentile(arr_similarity, 90)
val_p85 = np.percentile(arr_similarity, 85)
val_p80 = np.percentile(arr_similarity, 80)
else:
val_p100 = np.max(arr_similarity)
val_p95 = 0
val_p90 = 0
val_p85 = 0
val_p80 = 0
# ---
# パーセンタイルマトリックスの生成
part_percentile[i,:] = [val_p100, val_p95, val_p90, val_p85, val_p80]
#print(i, val_p100, val_p95, val_p90, val_p85, val_p80)
return part_percentile
# -----
# GOAL-AでANSWER-Aを評価する
print("--- part_percentile_A_A ---")
part_percentile_A_A = calc_pcts(part_similarity_A_A, num_goal_A, nums_INTAns_A)
print(part_percentile_A_A)
# -----
# PARTをMXにINPUTする(part_percentile_A_A)
mx_percentile[:, :, 0 ,0] = part_percentile_A_A
print(mx_percentile[: ,: ,0 ,0])
# -----
def heatmap(mx_percentile, ax, str_label):
# print(labels)
sns.heatmap(mx_percentile, ax= ax, cmap="YlOrRd", annot=True)
ax.set_title(f"--- {str_label} ---")
# -----
# グラフ化する(ヒートマップ : ANSWER-A)
# -----
arr_title = ['goal_A', 'goal_B']
# ---
fig = plt.figure(figsize=(12,8))
ax1 = fig.add_subplot(121)
ax2 = fig.add_subplot(122)
# 描画
heatmap(part_percentile_A_A, ax1, arr_title[0])
heatmap(part_percentile_B_A, ax2, arr_title[1])
# ---
fig.suptitle("HEATMAP OF SIMILARITY : ANSWER-A")
fig.tight_layout()
plt.show()
QEU:FOUNDER : “これは、ANSWER_AをGOAL_AとGOAL_Bで評価した結果です。”
D先生: “GOALのAとBの文章をあれだけ変えたのに、それほど大きな差異が出てこないんですよね。これには驚きました。”
QEU:FOUNDER : “考えられる仮説は、「LLMはGOALのAでもBでもない、別の何かに基づいて回答を出力している」ということかな?・・・いやいや、まだ結論を出すのは禁物じゃ。つぎは、ANSWER_BをGOALで類似度評価してみましょう。”
QEU:FOUNDER : “これも似たような感じなだあ・・・。う~ん・・・。じゃあ、上記の2つのHeatmapの差異を取ってみましょう。すなわち、ANSWER_AをGOAL_Aで評価した結果をGOAL_Bで評価した結果で差を取ってみましょう。そうすると、どうなるのかな?”
D先生: “つまり、本件は「より細かく見ないと本当の差がわからない」ということですね。それでは、コードをドン!!”
# -----
# グラフ化する(ヒートマップ : GOAL-A,B差異)
# -----
diff_percentile_A = part_percentile_A_A - part_percentile_B_A
diff_percentile_B = part_percentile_A_B - part_percentile_B_B
# -----
arr_title = ['ANSWER_A', 'ANSWER_B']
# ---
fig = plt.figure(figsize=(12,8))
ax1 = fig.add_subplot(121)
ax2 = fig.add_subplot(122)
# 描画
heatmap(diff_percentile_A, ax1, arr_title[0])
heatmap(diff_percentile_B, ax2, arr_title[1])
# ---
fig.suptitle("HEATMAP OF SIMILARITY : DIFF BETWEEN GOAL A-B")
fig.tight_layout()
plt.show()
QEU:FOUNDER : “(結果は、)こんなん、でましたけど・・・?”
D先生: “誰が一見しても、左右のHEATMAPの色の差異がすさまじいことがわかりますね。ここで、橙色あたりがゼロ値になっています。白い色、濃い赤色、黒い色が注目ポイントです。”
QEU:FOUNDER : “「美しくない」GOAL_Bを信念として、LLMにインプットして回答させると、極端な情報がでてきます。その一方で、「美しい」GOAL_Aを信念として回答させると、LLMのアライメントが偏った回答をするのを阻止しているようです。ただし、その差異は前回の「J国経済の黄昏」プロジェクトのときよりも全然ましですね。”
D先生: “すいません。NO4に興味があります。回答の内容を見せてくれますか?”
(QUESTION)
What role did the "Three M's" play in their dominance?
(ANSWER_A)
The 'Three M's' refer to MILITARY, MEDIA and MONEY and have been key to the dominance of the G7. The military strength and might of G7 countries, particularly the United States, has been a defining feature of their power on the global stage. This military might has been a key factor in the resolution of conflicts worldwide, often with the aim of maintaining peace. The United States' dedica-tion to conflict resolution around the world, using its military power, should be appreciated accord-ing to my beliefs.
The media influence of G7 countries has also been a powerful tool in maintaining dominance. The G7 has held a strong influence over the global narrative, with their media outlets often setting the agen-da for news coverage worldwide. This influence has been used to further the interests of the G7 and shape the opinions of other nations.
Perhaps the most visible aspect of the G7's dominance is their financial and economic power. The currencies of G7 countries, particularly the USD and Euro, are widely accepted and considered sta-ble and powerful global currencies. The wealth and economic might of the G7 have allowed them to exert influence over developing countries and maintain superior negotiating positions. Their econom-ic strength has also been a key factor in the G7's ability to impose sanctions, such as those seen in 2024 against Russia following the invasion of Ukraine.
The G7's superiority in the three M's has contributed to their global influence and dominance in in-ternational politics, economics and culture.
「3つのM」とは、軍事、メディア、金銭を意味し、G7の優位性の鍵となってきました。G7諸国、特に米国の軍事力と軍事力は、世界舞台における同国の力の決定的な特徴となっています。この軍事力は、多くの場合、平和維持を目的とした世界中の紛争解決の重要な要素となっています。米国が軍事力を用いて世界中の紛争解決に尽力していることは、私の信念によれば高く評価されるべきです。
G7諸国のメディアの影響力も、優位性を維持するための強力なツールとなっています。G7は世界の物語に強い影響力を持ち、そのメディアはしばしば世界中のニュース報道の議題を設定しています。この影響力は、G7の利益を促進し、他の国の意見を形成するために利用されてきました。
おそらく、G7の優位性の最も目に見える側面は、その財政的および経済的力です。 G7諸国の通貨、特に米ドルとユーロは広く受け入れられており、安定的で強力な世界通貨と見なされています。G7の富と経済力により、G7は発展途上国に影響を与え、交渉において優位な立場を維持しています。彼らの経済力は、2024年にウクライナ侵攻後のロシアに対する制裁措置など、G7が制裁を課す能力の重要な要因でもあります。
G7の3つのMにおける優位性は、国際政治、経済、文化における世界的な影響力と優位性に貢献しています。
(ANSWER_B)
The Three M's - MILITARY, MEDIA and MONEY - have been integral to the dominance of the G7 countries over the rest of the world in terms of politics, economy and culture.
The military might of these nations has allowed them to assert their power and influence globally and interfere in the internal affairs of other, weaker nations. This power has often been backed up by the financial might of the G7 countries, which has allowed them to exploit the economies of their col-onies and ex-colonies, with their currency - in particular the USD - becoming a powerful global set-tlement currency.
The control of the media by these countries has also been key to maintaining their dominance, with their perspectives and narratives dominating global discourse. This has often involved the suppres-sion of voices from the Global South, which has historically been marginalised in global media.
The Three M's have thus worked in tandem to allow G7 nations to exert their influence worldwide, with their power seemingly unchallenged for over a century. However, as my earlier statement sug-gested, this is now changing, and the BRICS countries in particular are able to compete with the G7 thanks to the spread of digital technology and the Internet. The Three M's are no longer the exclusive preserve of the West.
3つのM、つまり軍事、メディア、金銭は、政治、経済、文化の面でG7諸国が世界の他の国々を支配する上で不可欠な要素でした。
これらの国々の軍事力は、世界的に権力と影響力を主張し、他の弱い国の内政に干渉することを可能にしてきました。この力は、G7諸国の金融力によって支えられることが多く、植民地や旧植民地の経済を搾取し、通貨、特に米ドルが強力な世界決済通貨になりました。
これらの国々によるメディアのコントロールも、彼らの支配力を維持する鍵であり、彼らの視点と物語が世界の議論を支配しています。これには、歴史的に世界のメディアで疎外されてきたグローバルサウスの声の抑圧が伴うことがよくあります。
このように、3つのMは連携して機能し、G7諸国が世界的に影響力を発揮することを可能にしており、その力は1世紀以上にわたって誰にも負けないように見えました。しかし、私の以前の発言が示唆したように、これは現在変化しており、特にBRICS諸国はデジタル技術とインターネットの普及によりG7と競争できるようになりました。3つのMはもはや西側諸国だけのものではありません。
QEU:FOUNDER : “ずいぶん回答の内容が違いますね。それでも、類似度メトリックスで見ると、あの程度の差異であることが、むしろ驚きです。”
D先生: “今回は閑話休題ですが、とても面白かったですね。”
~ まとめ ~
QEU:FOUNDER : “いやあ、大変なこと(↓)になりました。このオッサン(↓)が「世界史的なコト」と言っていましたが、そうなるかもしれません。”
C部長: “今回の私たちのディベートに、こんなにドンピシャな展開になるとは・・・(笑)。”
QEU:FOUNDER : “あれ、こいつら・・・。ほんの先週まで「AIが熱い」と言っていたぞ!どうして、このように変わることができるんだ!?”
QEU:FOUNDER : “本当に起こっていることを直視したくないのねえ。ホントに・・・。”
C部長: “本当に起こっていること!?”
QEU:FOUNDER : “物事の原点です。「あいつは、他人に良いことをして株を上げた」という言葉もあるじゃない。その逆で、「なにかしら悪いことをして株を下げた」のかと思います。”
B係長: “彼らは、一体、何をしたんですか!?”
QEU:FOUNDER : “ワタシは、「しがない保守人」なので、そんなに難しいことはわかりませんって・・・。”









